Can I ask a question of Gbrowse2?

classic Classic list List threaded Threaded
15 messages Options
Reply | Threaded
Open this post in threaded view
|

Can I ask a question of Gbrowse2?

Jiantao Yu
Dear Sir/Madam,

I highly appreciate the smart of GBrowse. When I tried to display the quantitative data, and set up the zoom with small values, such as 40Kbp, I got the picture as is attached (picture1.png). However, I perfer the blocks to locate on the same line, rather than being hierarchical. I mean, It would be better if the blocks are kept as they are in the zoom of 500Kbp (all of the blocks are at one line, shown in the picture2.png). Could you please tell me how to do this? Many thanks!

BTW: my configure script is like this:

[methySW_MSH1_1000K]
feature         = methySW_MSH1_1000K
glyph           = xyplot
graph_type      = boxes
fgcolor         = red
bgcolor         = red
height          = 80
min_score       = 0
max_score       = 1
scale           = none
group_on        = methySW_MSH1_1000K
category        = Quantitative Data
label           = 0
key             = Methylation Profile - MSH1_dr_vs_Col0 (Sliding Windows:1000K-size)


Best wishes,
Jiantao

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Gmod-gbrowse mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse

Picture1.png (167K) Download Attachment
Picture2.png (126K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: [Gmod-help] Can I ask a question of Gbrowse2?

Scott Cain
Hello Jiantao,

What you are seeing is almost always a result of the config and the
data not matching.  I think your "group_on" setting is incorrect; it
is usually something like "group_on = display_name" or "group_on =
subject".  I don't think setting it equal to the name of the feature
type does anything.  If fixing that doesn't work, you could send a
sample of the GFF to the list.  Also, there is a section of the
tutorial that came with GBrowse that covers xyplots that might help
(look for "quantitative data" in the tutorial).

Scott


On Mon, Apr 23, 2012 at 8:14 PM, Jiantao Yu <[hidden email]> wrote:

> Dear Sir/Madam,
>
> I highly appreciate the smart of GBrowse. When I tried to display the
> quantitative data, and set up the zoom with small values, such as 40Kbp, I
> got the picture as is attached (picture1.png). However, I perfer the blocks
> to locate on the same line, rather than being hierarchical. I mean, It would
> be better if the blocks are kept as they are in the zoom of 500Kbp (all of
> the blocks are at one line, shown in the picture2.png). Could you please
> tell me how to do this? Many thanks!
>
> BTW: my configure script is like this:
>
> [methySW_MSH1_1000K]
> feature         = methySW_MSH1_1000K
> glyph           = xyplot
> graph_type      = boxes
> fgcolor         = red
> bgcolor         = red
> height          = 80
> min_score       = 0
> max_score       = 1
> scale           = none
> group_on        = methySW_MSH1_1000K
> category        = Quantitative Data
> label           = 0
> key             = Methylation Profile - MSH1_dr_vs_Col0 (Sliding
> Windows:1000K-size)
>
>
> Best wishes,
> Jiantao



--
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     216-392-3087
Ontario Institute for Cancer Research

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Gmod-gbrowse mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse
Reply | Threaded
Open this post in threaded view
|

Re: [Gmod-help] Can I ask a question of Gbrowse2?

Jiantao Yu
Hi Scott,

Really thanks for you quick reply. I tried to change group-on in the config file into
group_on = display_name                    or
group_on =subject

However, it seems no effect. I have attached the .gff3 and .conf files, and the configure script for the .gff3 data starts at the line of No.1365.

Could you please have a look at them? Thanks very much.

Best regards
Jiantao

2012/4/24 Scott Cain <[hidden email]>
Hello Jiantao,

What you are seeing is almost always a result of the config and the
data not matching.  I think your "group_on" setting is incorrect; it
is usually something like "group_on = display_name" or "group_on =
subject".  I don't think setting it equal to the name of the feature
type does anything.  If fixing that doesn't work, you could send a
sample of the GFF to the list.  Also, there is a section of the
tutorial that came with GBrowse that covers xyplots that might help
(look for "quantitative data" in the tutorial).

Scott


On Mon, Apr 23, 2012 at 8:14 PM, Jiantao Yu <[hidden email]> wrote:
> Dear Sir/Madam,
>
> I highly appreciate the smart of GBrowse. When I tried to display the
> quantitative data, and set up the zoom with small values, such as 40Kbp, I
> got the picture as is attached (picture1.png). However, I perfer the blocks
> to locate on the same line, rather than being hierarchical. I mean, It would
> be better if the blocks are kept as they are in the zoom of 500Kbp (all of
> the blocks are at one line, shown in the picture2.png). Could you please
> tell me how to do this? Many thanks!
>
> BTW: my configure script is like this:
>
> [methySW_MSH1_1000K]
> feature         = methySW_MSH1_1000K
> glyph           = xyplot
> graph_type      = boxes
> fgcolor         = red
> bgcolor         = red
> height          = 80
> min_score       = 0
> max_score       = 1
> scale           = none
> group_on        = methySW_MSH1_1000K
> category        = Quantitative Data
> label           = 0
> key             = Methylation Profile - MSH1_dr_vs_Col0 (Sliding
> Windows:1000K-size)
>
>
> Best wishes,
> Jiantao



--
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087">216-392-3087
Ontario Institute for Cancer Research


------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Gmod-gbrowse mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse

source.gff3 (3K) Download Attachment
a_thaliana.conf (74K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: [Gmod-help] Can I ask a question of Gbrowse2?

Scott Cain
Hi Jiantao,

"subject" should work for your data.  Did you try panning or zooming
to another location to make sure GBrowse wasn't using a cached image
(just reloading isn't enough)?

Scott


On Tue, Apr 24, 2012 at 11:44 AM, Jiantao Yu <[hidden email]> wrote:

> Hi Scott,
>
> Really thanks for you quick reply. I tried to change group-on in the config
> file into
> group_on = display_name                    or
> group_on =subject
>
> However, it seems no effect. I have attached the .gff3 and .conf files, and
> the configure script for the .gff3 data starts at the line of No.1365.
>
> Could you please have a look at them? Thanks very much.
>
> Best regards
> Jiantao
>
>
> 2012/4/24 Scott Cain <[hidden email]>
>>
>> Hello Jiantao,
>>
>> What you are seeing is almost always a result of the config and the
>> data not matching.  I think your "group_on" setting is incorrect; it
>> is usually something like "group_on = display_name" or "group_on =
>> subject".  I don't think setting it equal to the name of the feature
>> type does anything.  If fixing that doesn't work, you could send a
>> sample of the GFF to the list.  Also, there is a section of the
>> tutorial that came with GBrowse that covers xyplots that might help
>> (look for "quantitative data" in the tutorial).
>>
>> Scott
>>
>>
>> On Mon, Apr 23, 2012 at 8:14 PM, Jiantao Yu <[hidden email]>
>> wrote:
>> > Dear Sir/Madam,
>> >
>> > I highly appreciate the smart of GBrowse. When I tried to display the
>> > quantitative data, and set up the zoom with small values, such as 40Kbp,
>> > I
>> > got the picture as is attached (picture1.png). However, I perfer the
>> > blocks
>> > to locate on the same line, rather than being hierarchical. I mean, It
>> > would
>> > be better if the blocks are kept as they are in the zoom of 500Kbp (all
>> > of
>> > the blocks are at one line, shown in the picture2.png). Could you please
>> > tell me how to do this? Many thanks!
>> >
>> > BTW: my configure script is like this:
>> >
>> > [methySW_MSH1_1000K]
>> > feature         = methySW_MSH1_1000K
>> > glyph           = xyplot
>> > graph_type      = boxes
>> > fgcolor         = red
>> > bgcolor         = red
>> > height          = 80
>> > min_score       = 0
>> > max_score       = 1
>> > scale           = none
>> > group_on        = methySW_MSH1_1000K
>> > category        = Quantitative Data
>> > label           = 0
>> > key             = Methylation Profile - MSH1_dr_vs_Col0 (Sliding
>> > Windows:1000K-size)
>> >
>> >
>> > Best wishes,
>> > Jiantao
>>
>>
>>
>> --
>> ------------------------------------------------------------------------
>> Scott Cain, Ph. D.                                   scott at scottcain
>> dot net
>> GMOD Coordinator (http://gmod.org/)                     216-392-3087
>> Ontario Institute for Cancer Research
>
>



--
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     216-392-3087
Ontario Institute for Cancer Research

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Gmod-gbrowse mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse
Reply | Threaded
Open this post in threaded view
|

Re: [Gmod-help] Can I ask a question of Gbrowse2?

Scott Cain
Jiantao,

Please always respond to the mailing list too (reply-all).

No, you absolutely should not try to group these features by ID.  That
will result in the database loader putting all of the features
together into one gigantic object in the database, and your
performance will be very poor (GBrowse may stop rendering those tracks
altogether).  I'll try to reproduce your problem now.

Scott


On Tue, Apr 24, 2012 at 12:23 PM, Jiantao Yu <[hidden email]> wrote:

> Hi Scott,
>
> I still got the picture like the attahed file, although I changed
> group_on=subject. Do you think I need to add 'ID=1' for all of the entries
> in .gff3 file, in order to let them displayed on the single line?
>
> GBrowse may group entries according to their IDs? However, when I did this,
> I can't get the specific balloon for each position/block in the picture, you
> see, I just got a very general description in the balloon, such as this
> position is from 1..30Mbp, which is not what I want to get.
>
> Thanks a lot.
>
>
> Jiantao
>
> 2012/4/24 Scott Cain <[hidden email]>
>>
>> Hi Jiantao,
>>
>> "subject" should work for your data.  Did you try panning or zooming
>> to another location to make sure GBrowse wasn't using a cached image
>> (just reloading isn't enough)?
>>
>> Scott
>>
>>
>> On Tue, Apr 24, 2012 at 11:44 AM, Jiantao Yu <[hidden email]>
>> wrote:
>> > Hi Scott,
>> >
>> > Really thanks for you quick reply. I tried to change group-on in the
>> > config
>> > file into
>> > group_on = display_name                    or
>> > group_on =subject
>> >
>> > However, it seems no effect. I have attached the .gff3 and .conf files,
>> > and
>> > the configure script for the .gff3 data starts at the line of No.1365.
>> >
>> > Could you please have a look at them? Thanks very much.
>> >
>> > Best regards
>> > Jiantao
>> >
>> >
>> > 2012/4/24 Scott Cain <[hidden email]>
>> >>
>> >> Hello Jiantao,
>> >>
>> >> What you are seeing is almost always a result of the config and the
>> >> data not matching.  I think your "group_on" setting is incorrect; it
>> >> is usually something like "group_on = display_name" or "group_on =
>> >> subject".  I don't think setting it equal to the name of the feature
>> >> type does anything.  If fixing that doesn't work, you could send a
>> >> sample of the GFF to the list.  Also, there is a section of the
>> >> tutorial that came with GBrowse that covers xyplots that might help
>> >> (look for "quantitative data" in the tutorial).
>> >>
>> >> Scott
>> >>
>> >>
>> >> On Mon, Apr 23, 2012 at 8:14 PM, Jiantao Yu
>> >> <[hidden email]>
>> >> wrote:
>> >> > Dear Sir/Madam,
>> >> >
>> >> > I highly appreciate the smart of GBrowse. When I tried to display the
>> >> > quantitative data, and set up the zoom with small values, such as
>> >> > 40Kbp,
>> >> > I
>> >> > got the picture as is attached (picture1.png). However, I perfer the
>> >> > blocks
>> >> > to locate on the same line, rather than being hierarchical. I mean,
>> >> > It
>> >> > would
>> >> > be better if the blocks are kept as they are in the zoom of 500Kbp
>> >> > (all
>> >> > of
>> >> > the blocks are at one line, shown in the picture2.png). Could you
>> >> > please
>> >> > tell me how to do this? Many thanks!
>> >> >
>> >> > BTW: my configure script is like this:
>> >> >
>> >> > [methySW_MSH1_1000K]
>> >> > feature         = methySW_MSH1_1000K
>> >> > glyph           = xyplot
>> >> > graph_type      = boxes
>> >> > fgcolor         = red
>> >> > bgcolor         = red
>> >> > height          = 80
>> >> > min_score       = 0
>> >> > max_score       = 1
>> >> > scale           = none
>> >> > group_on        = methySW_MSH1_1000K
>> >> > category        = Quantitative Data
>> >> > label           = 0
>> >> > key             = Methylation Profile - MSH1_dr_vs_Col0 (Sliding
>> >> > Windows:1000K-size)
>> >> >
>> >> >
>> >> > Best wishes,
>> >> > Jiantao
>> >>
>> >>
>> >>
>> >> --
>> >>
>> >> ------------------------------------------------------------------------
>> >> Scott Cain, Ph. D.                                   scott at scottcain
>> >> dot net
>> >> GMOD Coordinator (http://gmod.org/)                     216-392-3087
>> >> Ontario Institute for Cancer Research
>> >
>> >
>>
>>
>>
>> --
>> ------------------------------------------------------------------------
>> Scott Cain, Ph. D.                                   scott at scottcain
>> dot net
>> GMOD Coordinator (http://gmod.org/)                     216-392-3087
>> Ontario Institute for Cancer Research
>
>



--
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     216-392-3087
Ontario Institute for Cancer Research

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Gmod-gbrowse mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse
Reply | Threaded
Open this post in threaded view
|

Re: [Gmod-help] Can I ask a question of Gbrowse2?

Jiantao Yu
Hi Scott,

Thanks for reply. I found some answers for the three questions:
i) The reason that the boxes are displayed in hierarchy, I think, would be the boxes are too close. For example, if the start and end positions are like this:

1    10    0.8    .    .   
11    20    0.7    .    .   
21    30    0.7    .    .   
...

They would probably be in hierarchy when I zoom in. However, if I changed the positions like this (+ or - 4):

4    7    0.8    .    .   
14    17    0.7    .    .   
24    27    0.7    .    .   

They would be on the same line. However, in the case of my data, it is hard to determine what values should be added or subtracted. For example, I used 30 as the value, it wasn't enough, I mean, 30 is not enough to ensure all of the boxes are on the same line for every scale of zooming in or zooming out.

From another point, I really hope the width of boxes are still kept continuous, you see, no gaps.

ii) I tried to gave all of the entries, located in the same chromosome, the same IDs. This method did solve the problem, i.e. the boxes are fixed on the same line. However, as you told in the last email, this led to a problem, the performance became poor, and the gbrowse couldn't display the curves sometimes, if there were lots of boxes which had the same IDs.

So I think the best way would be, can we find a way to display them on the same line, while no limitation of their IDs?

iii) The reason that there is no title of some tracks, is not very clear. However, when I changed the 'key' with some other words, the title can be normally displayed. So this has not been a problem now.

Thanks very much for your patience.

Regards
Jiantao


2012/4/27 Scott Cain <[hidden email]>
Jiantao,

Please always respond to the mailing list too (reply-all).

No, you absolutely should not try to group these features by ID.  That
will result in the database loader putting all of the features
together into one gigantic object in the database, and your
performance will be very poor (GBrowse may stop rendering those tracks
altogether).  I'll try to reproduce your problem now.

Scott


On Tue, Apr 24, 2012 at 12:23 PM, Jiantao Yu <[hidden email]> wrote:
> Hi Scott,
>
> I still got the picture like the attahed file, although I changed
> group_on=subject. Do you think I need to add 'ID=1' for all of the entries
> in .gff3 file, in order to let them displayed on the single line?
>
> GBrowse may group entries according to their IDs? However, when I did this,
> I can't get the specific balloon for each position/block in the picture, you
> see, I just got a very general description in the balloon, such as this
> position is from 1..30Mbp, which is not what I want to get.
>
> Thanks a lot.
>
>
> Jiantao
>
> 2012/4/24 Scott Cain <[hidden email]>
>>
>> Hi Jiantao,
>>
>> "subject" should work for your data.  Did you try panning or zooming
>> to another location to make sure GBrowse wasn't using a cached image
>> (just reloading isn't enough)?
>>
>> Scott
>>
>>
>> On Tue, Apr 24, 2012 at 11:44 AM, Jiantao Yu <[hidden email]>
>> wrote:
>> > Hi Scott,
>> >
>> > Really thanks for you quick reply. I tried to change group-on in the
>> > config
>> > file into
>> > group_on = display_name                    or
>> > group_on =subject
>> >
>> > However, it seems no effect. I have attached the .gff3 and .conf files,
>> > and
>> > the configure script for the .gff3 data starts at the line of No.1365.
>> >
>> > Could you please have a look at them? Thanks very much.
>> >
>> > Best regards
>> > Jiantao
>> >
>> >
>> > 2012/4/24 Scott Cain <[hidden email]>
>> >>
>> >> Hello Jiantao,
>> >>
>> >> What you are seeing is almost always a result of the config and the
>> >> data not matching.  I think your "group_on" setting is incorrect; it
>> >> is usually something like "group_on = display_name" or "group_on =
>> >> subject".  I don't think setting it equal to the name of the feature
>> >> type does anything.  If fixing that doesn't work, you could send a
>> >> sample of the GFF to the list.  Also, there is a section of the
>> >> tutorial that came with GBrowse that covers xyplots that might help
>> >> (look for "quantitative data" in the tutorial).
>> >>
>> >> Scott
>> >>
>> >>
>> >> On Mon, Apr 23, 2012 at 8:14 PM, Jiantao Yu
>> >> <[hidden email]>
>> >> wrote:
>> >> > Dear Sir/Madam,
>> >> >
>> >> > I highly appreciate the smart of GBrowse. When I tried to display the
>> >> > quantitative data, and set up the zoom with small values, such as
>> >> > 40Kbp,
>> >> > I
>> >> > got the picture as is attached (picture1.png). However, I perfer the
>> >> > blocks
>> >> > to locate on the same line, rather than being hierarchical. I mean,
>> >> > It
>> >> > would
>> >> > be better if the blocks are kept as they are in the zoom of 500Kbp
>> >> > (all
>> >> > of
>> >> > the blocks are at one line, shown in the picture2.png). Could you
>> >> > please
>> >> > tell me how to do this? Many thanks!
>> >> >
>> >> > BTW: my configure script is like this:
>> >> >
>> >> > [methySW_MSH1_1000K]
>> >> > feature         = methySW_MSH1_1000K
>> >> > glyph           = xyplot
>> >> > graph_type      = boxes
>> >> > fgcolor         = red
>> >> > bgcolor         = red
>> >> > height          = 80
>> >> > min_score       = 0
>> >> > max_score       = 1
>> >> > scale           = none
>> >> > group_on        = methySW_MSH1_1000K
>> >> > category        = Quantitative Data
>> >> > label           = 0
>> >> > key             = Methylation Profile - MSH1_dr_vs_Col0 (Sliding
>> >> > Windows:1000K-size)
>> >> >
>> >> >
>> >> > Best wishes,
>> >> > Jiantao
>> >>
>> >>
>> >>
>> >> --
>> >>
>> >> ------------------------------------------------------------------------
>> >> Scott Cain, Ph. D.                                   scott at scottcain
>> >> dot net
>> >> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087">216-392-3087
>> >> Ontario Institute for Cancer Research
>> >
>> >
>>
>>
>>
>> --
>> ------------------------------------------------------------------------
>> Scott Cain, Ph. D.                                   scott at scottcain
>> dot net
>> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087">216-392-3087
>> Ontario Institute for Cancer Research
>
>



--
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087">216-392-3087
Ontario Institute for Cancer Research


------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Gmod-gbrowse mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse
Reply | Threaded
Open this post in threaded view
|

Re: [Gmod-help] Can I ask a question of Gbrowse2?

Scott Cain
Hi Jiantao,

I forgot to ask--what version of GBrowse are you using?  Your conf
file looks like GBrowse 1.X.  Also, I trimmed the GMOD help desk email
off the cc list.

For xyplots: having the boundaries agacent shouldn't matter; that is
what the data in the tutorial are, and it works fine.  There is a
problem either with your data, your config or both.  I'm working on a
simplified case to see if I can reproduce your problem.

For the key/track title problem, I'm guessing you had illigal characters.

Scott


On Fri, Apr 27, 2012 at 12:13 PM, Jiantao Yu <[hidden email]> wrote:

> Hi Scott,
>
> Thanks for reply. I found some answers for the three questions:
> i) The reason that the boxes are displayed in hierarchy, I think, would be
> the boxes are too close. For example, if the start and end positions are
> like this:
>
> 1    10    0.8    .    .
> 11    20    0.7    .    .
> 21    30    0.7    .    .
> ...
>
> They would probably be in hierarchy when I zoom in. However, if I changed
> the positions like this (+ or - 4):
>
> 4    7    0.8    .    .
> 14    17    0.7    .    .
> 24    27    0.7    .    .
>
> They would be on the same line. However, in the case of my data, it is hard
> to determine what values should be added or subtracted. For example, I used
> 30 as the value, it wasn't enough, I mean, 30 is not enough to ensure all of
> the boxes are on the same line for every scale of zooming in or zooming out.
>
> From another point, I really hope the width of boxes are still kept
> continuous, you see, no gaps.
>
> ii) I tried to gave all of the entries, located in the same chromosome, the
> same IDs. This method did solve the problem, i.e. the boxes are fixed on the
> same line. However, as you told in the last email, this led to a problem,
> the performance became poor, and the gbrowse couldn't display the curves
> sometimes, if there were lots of boxes which had the same IDs.
>
> So I think the best way would be, can we find a way to display them on the
> same line, while no limitation of their IDs?
>
> iii) The reason that there is no title of some tracks, is not very clear.
> However, when I changed the 'key' with some other words, the title can be
> normally displayed. So this has not been a problem now.
>
> Thanks very much for your patience.
>
> Regards
> Jiantao
>
>
>
> 2012/4/27 Scott Cain <[hidden email]>
>>
>> Jiantao,
>>
>> Please always respond to the mailing list too (reply-all).
>>
>> No, you absolutely should not try to group these features by ID.  That
>> will result in the database loader putting all of the features
>> together into one gigantic object in the database, and your
>> performance will be very poor (GBrowse may stop rendering those tracks
>> altogether).  I'll try to reproduce your problem now.
>>
>> Scott
>>
>>
>> On Tue, Apr 24, 2012 at 12:23 PM, Jiantao Yu <[hidden email]>
>> wrote:
>> > Hi Scott,
>> >
>> > I still got the picture like the attahed file, although I changed
>> > group_on=subject. Do you think I need to add 'ID=1' for all of the
>> > entries
>> > in .gff3 file, in order to let them displayed on the single line?
>> >
>> > GBrowse may group entries according to their IDs? However, when I did
>> > this,
>> > I can't get the specific balloon for each position/block in the picture,
>> > you
>> > see, I just got a very general description in the balloon, such as this
>> > position is from 1..30Mbp, which is not what I want to get.
>> >
>> > Thanks a lot.
>> >
>> >
>> > Jiantao
>> >
>> > 2012/4/24 Scott Cain <[hidden email]>
>> >>
>> >> Hi Jiantao,
>> >>
>> >> "subject" should work for your data.  Did you try panning or zooming
>> >> to another location to make sure GBrowse wasn't using a cached image
>> >> (just reloading isn't enough)?
>> >>
>> >> Scott
>> >>
>> >>
>> >> On Tue, Apr 24, 2012 at 11:44 AM, Jiantao Yu
>> >> <[hidden email]>
>> >> wrote:
>> >> > Hi Scott,
>> >> >
>> >> > Really thanks for you quick reply. I tried to change group-on in the
>> >> > config
>> >> > file into
>> >> > group_on = display_name                    or
>> >> > group_on =subject
>> >> >
>> >> > However, it seems no effect. I have attached the .gff3 and .conf
>> >> > files,
>> >> > and
>> >> > the configure script for the .gff3 data starts at the line of
>> >> > No.1365.
>> >> >
>> >> > Could you please have a look at them? Thanks very much.
>> >> >
>> >> > Best regards
>> >> > Jiantao
>> >> >
>> >> >
>> >> > 2012/4/24 Scott Cain <[hidden email]>
>> >> >>
>> >> >> Hello Jiantao,
>> >> >>
>> >> >> What you are seeing is almost always a result of the config and the
>> >> >> data not matching.  I think your "group_on" setting is incorrect; it
>> >> >> is usually something like "group_on = display_name" or "group_on =
>> >> >> subject".  I don't think setting it equal to the name of the feature
>> >> >> type does anything.  If fixing that doesn't work, you could send a
>> >> >> sample of the GFF to the list.  Also, there is a section of the
>> >> >> tutorial that came with GBrowse that covers xyplots that might help
>> >> >> (look for "quantitative data" in the tutorial).
>> >> >>
>> >> >> Scott
>> >> >>
>> >> >>
>> >> >> On Mon, Apr 23, 2012 at 8:14 PM, Jiantao Yu
>> >> >> <[hidden email]>
>> >> >> wrote:
>> >> >> > Dear Sir/Madam,
>> >> >> >
>> >> >> > I highly appreciate the smart of GBrowse. When I tried to display
>> >> >> > the
>> >> >> > quantitative data, and set up the zoom with small values, such as
>> >> >> > 40Kbp,
>> >> >> > I
>> >> >> > got the picture as is attached (picture1.png). However, I perfer
>> >> >> > the
>> >> >> > blocks
>> >> >> > to locate on the same line, rather than being hierarchical. I
>> >> >> > mean,
>> >> >> > It
>> >> >> > would
>> >> >> > be better if the blocks are kept as they are in the zoom of 500Kbp
>> >> >> > (all
>> >> >> > of
>> >> >> > the blocks are at one line, shown in the picture2.png). Could you
>> >> >> > please
>> >> >> > tell me how to do this? Many thanks!
>> >> >> >
>> >> >> > BTW: my configure script is like this:
>> >> >> >
>> >> >> > [methySW_MSH1_1000K]
>> >> >> > feature         = methySW_MSH1_1000K
>> >> >> > glyph           = xyplot
>> >> >> > graph_type      = boxes
>> >> >> > fgcolor         = red
>> >> >> > bgcolor         = red
>> >> >> > height          = 80
>> >> >> > min_score       = 0
>> >> >> > max_score       = 1
>> >> >> > scale           = none
>> >> >> > group_on        = methySW_MSH1_1000K
>> >> >> > category        = Quantitative Data
>> >> >> > label           = 0
>> >> >> > key             = Methylation Profile - MSH1_dr_vs_Col0 (Sliding
>> >> >> > Windows:1000K-size)
>> >> >> >
>> >> >> >
>> >> >> > Best wishes,
>> >> >> > Jiantao
>> >> >>
>> >> >>
>> >> >>
>> >> >> --
>> >> >>
>> >> >>
>> >> >> ------------------------------------------------------------------------
>> >> >> Scott Cain, Ph. D.                                   scott at
>> >> >> scottcain
>> >> >> dot net
>> >> >> GMOD Coordinator (http://gmod.org/)                     216-392-3087
>> >> >> Ontario Institute for Cancer Research
>> >> >
>> >> >
>> >>
>> >>
>> >>
>> >> --
>> >>
>> >> ------------------------------------------------------------------------
>> >> Scott Cain, Ph. D.                                   scott at scottcain
>> >> dot net
>> >> GMOD Coordinator (http://gmod.org/)                     216-392-3087
>> >> Ontario Institute for Cancer Research
>> >
>> >
>>
>>
>>
>> --
>> ------------------------------------------------------------------------
>> Scott Cain, Ph. D.                                   scott at scottcain
>> dot net
>> GMOD Coordinator (http://gmod.org/)                     216-392-3087
>> Ontario Institute for Cancer Research
>
>



--
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     216-392-3087
Ontario Institute for Cancer Research

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Gmod-gbrowse mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse
Reply | Threaded
Open this post in threaded view
|

Re: [Gmod-help] Can I ask a question of Gbrowse2?

Jiantao Yu
Hi Scott,

I think I am using Gbrowse2, and I made the configure file based on your instruction file, Generic Genome Browser Version 2: A Tutorial for Administrators.

However, I am not 100 percent sure of this. Can you tell me how can I check the version of Gbrowse.

Another point is that, you see, GBrowse2 didn't provide the configure file of Arabidopsis, so we generated this file by ourselves. However, the file is made by my colleague, so I am not sure if he took this configure file from the GBrowse1.x or Gbrowse2.x . If you are sure the configure file is from GBrowse1.x, I will ask my colleague and make sure of this. Thanks.

Jiantao

2012/4/27 Scott Cain <[hidden email]>
Hi Jiantao,

I forgot to ask--what version of GBrowse are you using?  Your conf
file looks like GBrowse 1.X.  Also, I trimmed the GMOD help desk email
off the cc list.

For xyplots: having the boundaries agacent shouldn't matter; that is
what the data in the tutorial are, and it works fine.  There is a
problem either with your data, your config or both.  I'm working on a
simplified case to see if I can reproduce your problem.

For the key/track title problem, I'm guessing you had illigal characters.

Scott


On Fri, Apr 27, 2012 at 12:13 PM, Jiantao Yu <[hidden email]> wrote:
> Hi Scott,
>
> Thanks for reply. I found some answers for the three questions:
> i) The reason that the boxes are displayed in hierarchy, I think, would be
> the boxes are too close. For example, if the start and end positions are
> like this:
>
> 1    10    0.8    .    .
> 11    20    0.7    .    .
> 21    30    0.7    .    .
> ...
>
> They would probably be in hierarchy when I zoom in. However, if I changed
> the positions like this (+ or - 4):
>
> 4    7    0.8    .    .
> 14    17    0.7    .    .
> 24    27    0.7    .    .
>
> They would be on the same line. However, in the case of my data, it is hard
> to determine what values should be added or subtracted. For example, I used
> 30 as the value, it wasn't enough, I mean, 30 is not enough to ensure all of
> the boxes are on the same line for every scale of zooming in or zooming out.
>
> From another point, I really hope the width of boxes are still kept
> continuous, you see, no gaps.
>
> ii) I tried to gave all of the entries, located in the same chromosome, the
> same IDs. This method did solve the problem, i.e. the boxes are fixed on the
> same line. However, as you told in the last email, this led to a problem,
> the performance became poor, and the gbrowse couldn't display the curves
> sometimes, if there were lots of boxes which had the same IDs.
>
> So I think the best way would be, can we find a way to display them on the
> same line, while no limitation of their IDs?
>
> iii) The reason that there is no title of some tracks, is not very clear.
> However, when I changed the 'key' with some other words, the title can be
> normally displayed. So this has not been a problem now.
>
> Thanks very much for your patience.
>
> Regards
> Jiantao
>
>
>
> 2012/4/27 Scott Cain <[hidden email]>
>>
>> Jiantao,
>>
>> Please always respond to the mailing list too (reply-all).
>>
>> No, you absolutely should not try to group these features by ID.  That
>> will result in the database loader putting all of the features
>> together into one gigantic object in the database, and your
>> performance will be very poor (GBrowse may stop rendering those tracks
>> altogether).  I'll try to reproduce your problem now.
>>
>> Scott
>>
>>
>> On Tue, Apr 24, 2012 at 12:23 PM, Jiantao Yu <[hidden email]>
>> wrote:
>> > Hi Scott,
>> >
>> > I still got the picture like the attahed file, although I changed
>> > group_on=subject. Do you think I need to add 'ID=1' for all of the
>> > entries
>> > in .gff3 file, in order to let them displayed on the single line?
>> >
>> > GBrowse may group entries according to their IDs? However, when I did
>> > this,
>> > I can't get the specific balloon for each position/block in the picture,
>> > you
>> > see, I just got a very general description in the balloon, such as this
>> > position is from 1..30Mbp, which is not what I want to get.
>> >
>> > Thanks a lot.
>> >
>> >
>> > Jiantao
>> >
>> > 2012/4/24 Scott Cain <[hidden email]>
>> >>
>> >> Hi Jiantao,
>> >>
>> >> "subject" should work for your data.  Did you try panning or zooming
>> >> to another location to make sure GBrowse wasn't using a cached image
>> >> (just reloading isn't enough)?
>> >>
>> >> Scott
>> >>
>> >>
>> >> On Tue, Apr 24, 2012 at 11:44 AM, Jiantao Yu
>> >> <[hidden email]>
>> >> wrote:
>> >> > Hi Scott,
>> >> >
>> >> > Really thanks for you quick reply. I tried to change group-on in the
>> >> > config
>> >> > file into
>> >> > group_on = display_name                    or
>> >> > group_on =subject
>> >> >
>> >> > However, it seems no effect. I have attached the .gff3 and .conf
>> >> > files,
>> >> > and
>> >> > the configure script for the .gff3 data starts at the line of
>> >> > No.1365.
>> >> >
>> >> > Could you please have a look at them? Thanks very much.
>> >> >
>> >> > Best regards
>> >> > Jiantao
>> >> >
>> >> >
>> >> > 2012/4/24 Scott Cain <[hidden email]>
>> >> >>
>> >> >> Hello Jiantao,
>> >> >>
>> >> >> What you are seeing is almost always a result of the config and the
>> >> >> data not matching.  I think your "group_on" setting is incorrect; it
>> >> >> is usually something like "group_on = display_name" or "group_on =
>> >> >> subject".  I don't think setting it equal to the name of the feature
>> >> >> type does anything.  If fixing that doesn't work, you could send a
>> >> >> sample of the GFF to the list.  Also, there is a section of the
>> >> >> tutorial that came with GBrowse that covers xyplots that might help
>> >> >> (look for "quantitative data" in the tutorial).
>> >> >>
>> >> >> Scott
>> >> >>
>> >> >>
>> >> >> On Mon, Apr 23, 2012 at 8:14 PM, Jiantao Yu
>> >> >> <[hidden email]>
>> >> >> wrote:
>> >> >> > Dear Sir/Madam,
>> >> >> >
>> >> >> > I highly appreciate the smart of GBrowse. When I tried to display
>> >> >> > the
>> >> >> > quantitative data, and set up the zoom with small values, such as
>> >> >> > 40Kbp,
>> >> >> > I
>> >> >> > got the picture as is attached (picture1.png). However, I perfer
>> >> >> > the
>> >> >> > blocks
>> >> >> > to locate on the same line, rather than being hierarchical. I
>> >> >> > mean,
>> >> >> > It
>> >> >> > would
>> >> >> > be better if the blocks are kept as they are in the zoom of 500Kbp
>> >> >> > (all
>> >> >> > of
>> >> >> > the blocks are at one line, shown in the picture2.png). Could you
>> >> >> > please
>> >> >> > tell me how to do this? Many thanks!
>> >> >> >
>> >> >> > BTW: my configure script is like this:
>> >> >> >
>> >> >> > [methySW_MSH1_1000K]
>> >> >> > feature         = methySW_MSH1_1000K
>> >> >> > glyph           = xyplot
>> >> >> > graph_type      = boxes
>> >> >> > fgcolor         = red
>> >> >> > bgcolor         = red
>> >> >> > height          = 80
>> >> >> > min_score       = 0
>> >> >> > max_score       = 1
>> >> >> > scale           = none
>> >> >> > group_on        = methySW_MSH1_1000K
>> >> >> > category        = Quantitative Data
>> >> >> > label           = 0
>> >> >> > key             = Methylation Profile - MSH1_dr_vs_Col0 (Sliding
>> >> >> > Windows:1000K-size)
>> >> >> >
>> >> >> >
>> >> >> > Best wishes,
>> >> >> > Jiantao
>> >> >>
>> >> >>
>> >> >>
>> >> >> --
>> >> >>
>> >> >>
>> >> >> ------------------------------------------------------------------------
>> >> >> Scott Cain, Ph. D.                                   scott at
>> >> >> scottcain
>> >> >> dot net
>> >> >> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087">216-392-3087
>> >> >> Ontario Institute for Cancer Research
>> >> >
>> >> >
>> >>
>> >>
>> >>
>> >> --
>> >>
>> >> ------------------------------------------------------------------------
>> >> Scott Cain, Ph. D.                                   scott at scottcain
>> >> dot net
>> >> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087">216-392-3087
>> >> Ontario Institute for Cancer Research
>> >
>> >
>>
>>
>>
>> --
>> ------------------------------------------------------------------------
>> Scott Cain, Ph. D.                                   scott at scottcain
>> dot net
>> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087">216-392-3087
>> Ontario Institute for Cancer Research
>
>



--
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087">216-392-3087
Ontario Institute for Cancer Research


------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Gmod-gbrowse mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse
Reply | Threaded
Open this post in threaded view
|

Re: [Gmod-help] Can I ask a question of Gbrowse2?

Scott Cain
Hi Jiantao,

The easiest way to tell what version of GBrowse you are using is the
location of the config files: if they are int /etc/gbrowse, it's
GBrowse2.  If not, it's probably GBrowse 1.

Since part of the problem you are having is related to how your data
are arranged in one track, could you send me a set of real data for
one of those tracks rather than a small sample that don't really
represent the problem?

Thanks,
Scott


On Fri, Apr 27, 2012 at 12:51 PM, Jiantao Yu <[hidden email]> wrote:

> Hi Scott,
>
> I think I am using Gbrowse2, and I made the configure file based on your
> instruction file, Generic Genome Browser Version 2: A Tutorial for
> Administrators.
>
> However, I am not 100 percent sure of this. Can you tell me how can I check
> the version of Gbrowse.
>
> Another point is that, you see, GBrowse2 didn't provide the configure file
> of Arabidopsis, so we generated this file by ourselves. However, the file is
> made by my colleague, so I am not sure if he took this configure file from
> the GBrowse1.x or Gbrowse2.x . If you are sure the configure file is from
> GBrowse1.x, I will ask my colleague and make sure of this. Thanks.
>
>
> Jiantao
>
> 2012/4/27 Scott Cain <[hidden email]>
>>
>> Hi Jiantao,
>>
>> I forgot to ask--what version of GBrowse are you using?  Your conf
>> file looks like GBrowse 1.X.  Also, I trimmed the GMOD help desk email
>> off the cc list.
>>
>> For xyplots: having the boundaries agacent shouldn't matter; that is
>> what the data in the tutorial are, and it works fine.  There is a
>> problem either with your data, your config or both.  I'm working on a
>> simplified case to see if I can reproduce your problem.
>>
>> For the key/track title problem, I'm guessing you had illigal characters.
>>
>> Scott
>>
>>
>> On Fri, Apr 27, 2012 at 12:13 PM, Jiantao Yu <[hidden email]>
>> wrote:
>> > Hi Scott,
>> >
>> > Thanks for reply. I found some answers for the three questions:
>> > i) The reason that the boxes are displayed in hierarchy, I think, would
>> > be
>> > the boxes are too close. For example, if the start and end positions are
>> > like this:
>> >
>> > 1    10    0.8    .    .
>> > 11    20    0.7    .    .
>> > 21    30    0.7    .    .
>> > ...
>> >
>> > They would probably be in hierarchy when I zoom in. However, if I
>> > changed
>> > the positions like this (+ or - 4):
>> >
>> > 4    7    0.8    .    .
>> > 14    17    0.7    .    .
>> > 24    27    0.7    .    .
>> >
>> > They would be on the same line. However, in the case of my data, it is
>> > hard
>> > to determine what values should be added or subtracted. For example, I
>> > used
>> > 30 as the value, it wasn't enough, I mean, 30 is not enough to ensure
>> > all of
>> > the boxes are on the same line for every scale of zooming in or zooming
>> > out.
>> >
>> > From another point, I really hope the width of boxes are still kept
>> > continuous, you see, no gaps.
>> >
>> > ii) I tried to gave all of the entries, located in the same chromosome,
>> > the
>> > same IDs. This method did solve the problem, i.e. the boxes are fixed on
>> > the
>> > same line. However, as you told in the last email, this led to a
>> > problem,
>> > the performance became poor, and the gbrowse couldn't display the curves
>> > sometimes, if there were lots of boxes which had the same IDs.
>> >
>> > So I think the best way would be, can we find a way to display them on
>> > the
>> > same line, while no limitation of their IDs?
>> >
>> > iii) The reason that there is no title of some tracks, is not very
>> > clear.
>> > However, when I changed the 'key' with some other words, the title can
>> > be
>> > normally displayed. So this has not been a problem now.
>> >
>> > Thanks very much for your patience.
>> >
>> > Regards
>> > Jiantao
>> >
>> >
>> >
>> > 2012/4/27 Scott Cain <[hidden email]>
>> >>
>> >> Jiantao,
>> >>
>> >> Please always respond to the mailing list too (reply-all).
>> >>
>> >> No, you absolutely should not try to group these features by ID.  That
>> >> will result in the database loader putting all of the features
>> >> together into one gigantic object in the database, and your
>> >> performance will be very poor (GBrowse may stop rendering those tracks
>> >> altogether).  I'll try to reproduce your problem now.
>> >>
>> >> Scott
>> >>
>> >>
>> >> On Tue, Apr 24, 2012 at 12:23 PM, Jiantao Yu
>> >> <[hidden email]>
>> >> wrote:
>> >> > Hi Scott,
>> >> >
>> >> > I still got the picture like the attahed file, although I changed
>> >> > group_on=subject. Do you think I need to add 'ID=1' for all of the
>> >> > entries
>> >> > in .gff3 file, in order to let them displayed on the single line?
>> >> >
>> >> > GBrowse may group entries according to their IDs? However, when I did
>> >> > this,
>> >> > I can't get the specific balloon for each position/block in the
>> >> > picture,
>> >> > you
>> >> > see, I just got a very general description in the balloon, such as
>> >> > this
>> >> > position is from 1..30Mbp, which is not what I want to get.
>> >> >
>> >> > Thanks a lot.
>> >> >
>> >> >
>> >> > Jiantao
>> >> >
>> >> > 2012/4/24 Scott Cain <[hidden email]>
>> >> >>
>> >> >> Hi Jiantao,
>> >> >>
>> >> >> "subject" should work for your data.  Did you try panning or zooming
>> >> >> to another location to make sure GBrowse wasn't using a cached image
>> >> >> (just reloading isn't enough)?
>> >> >>
>> >> >> Scott
>> >> >>
>> >> >>
>> >> >> On Tue, Apr 24, 2012 at 11:44 AM, Jiantao Yu
>> >> >> <[hidden email]>
>> >> >> wrote:
>> >> >> > Hi Scott,
>> >> >> >
>> >> >> > Really thanks for you quick reply. I tried to change group-on in
>> >> >> > the
>> >> >> > config
>> >> >> > file into
>> >> >> > group_on = display_name                    or
>> >> >> > group_on =subject
>> >> >> >
>> >> >> > However, it seems no effect. I have attached the .gff3 and .conf
>> >> >> > files,
>> >> >> > and
>> >> >> > the configure script for the .gff3 data starts at the line of
>> >> >> > No.1365.
>> >> >> >
>> >> >> > Could you please have a look at them? Thanks very much.
>> >> >> >
>> >> >> > Best regards
>> >> >> > Jiantao
>> >> >> >
>> >> >> >
>> >> >> > 2012/4/24 Scott Cain <[hidden email]>
>> >> >> >>
>> >> >> >> Hello Jiantao,
>> >> >> >>
>> >> >> >> What you are seeing is almost always a result of the config and
>> >> >> >> the
>> >> >> >> data not matching.  I think your "group_on" setting is incorrect;
>> >> >> >> it
>> >> >> >> is usually something like "group_on = display_name" or "group_on
>> >> >> >> =
>> >> >> >> subject".  I don't think setting it equal to the name of the
>> >> >> >> feature
>> >> >> >> type does anything.  If fixing that doesn't work, you could send
>> >> >> >> a
>> >> >> >> sample of the GFF to the list.  Also, there is a section of the
>> >> >> >> tutorial that came with GBrowse that covers xyplots that might
>> >> >> >> help
>> >> >> >> (look for "quantitative data" in the tutorial).
>> >> >> >>
>> >> >> >> Scott
>> >> >> >>
>> >> >> >>
>> >> >> >> On Mon, Apr 23, 2012 at 8:14 PM, Jiantao Yu
>> >> >> >> <[hidden email]>
>> >> >> >> wrote:
>> >> >> >> > Dear Sir/Madam,
>> >> >> >> >
>> >> >> >> > I highly appreciate the smart of GBrowse. When I tried to
>> >> >> >> > display
>> >> >> >> > the
>> >> >> >> > quantitative data, and set up the zoom with small values, such
>> >> >> >> > as
>> >> >> >> > 40Kbp,
>> >> >> >> > I
>> >> >> >> > got the picture as is attached (picture1.png). However, I
>> >> >> >> > perfer
>> >> >> >> > the
>> >> >> >> > blocks
>> >> >> >> > to locate on the same line, rather than being hierarchical. I
>> >> >> >> > mean,
>> >> >> >> > It
>> >> >> >> > would
>> >> >> >> > be better if the blocks are kept as they are in the zoom of
>> >> >> >> > 500Kbp
>> >> >> >> > (all
>> >> >> >> > of
>> >> >> >> > the blocks are at one line, shown in the picture2.png). Could
>> >> >> >> > you
>> >> >> >> > please
>> >> >> >> > tell me how to do this? Many thanks!
>> >> >> >> >
>> >> >> >> > BTW: my configure script is like this:
>> >> >> >> >
>> >> >> >> > [methySW_MSH1_1000K]
>> >> >> >> > feature         = methySW_MSH1_1000K
>> >> >> >> > glyph           = xyplot
>> >> >> >> > graph_type      = boxes
>> >> >> >> > fgcolor         = red
>> >> >> >> > bgcolor         = red
>> >> >> >> > height          = 80
>> >> >> >> > min_score       = 0
>> >> >> >> > max_score       = 1
>> >> >> >> > scale           = none
>> >> >> >> > group_on        = methySW_MSH1_1000K
>> >> >> >> > category        = Quantitative Data
>> >> >> >> > label           = 0
>> >> >> >> > key             = Methylation Profile - MSH1_dr_vs_Col0
>> >> >> >> > (Sliding
>> >> >> >> > Windows:1000K-size)
>> >> >> >> >
>> >> >> >> >
>> >> >> >> > Best wishes,
>> >> >> >> > Jiantao
>> >> >> >>
>> >> >> >>
>> >> >> >>
>> >> >> >> --
>> >> >> >>
>> >> >> >>
>> >> >> >>
>> >> >> >> ------------------------------------------------------------------------
>> >> >> >> Scott Cain, Ph. D.                                   scott at
>> >> >> >> scottcain
>> >> >> >> dot net
>> >> >> >> GMOD Coordinator (http://gmod.org/)
>> >> >> >> 216-392-3087
>> >> >> >> Ontario Institute for Cancer Research
>> >> >> >
>> >> >> >
>> >> >>
>> >> >>
>> >> >>
>> >> >> --
>> >> >>
>> >> >>
>> >> >> ------------------------------------------------------------------------
>> >> >> Scott Cain, Ph. D.                                   scott at
>> >> >> scottcain
>> >> >> dot net
>> >> >> GMOD Coordinator (http://gmod.org/)                     216-392-3087
>> >> >> Ontario Institute for Cancer Research
>> >> >
>> >> >
>> >>
>> >>
>> >>
>> >> --
>> >>
>> >> ------------------------------------------------------------------------
>> >> Scott Cain, Ph. D.                                   scott at scottcain
>> >> dot net
>> >> GMOD Coordinator (http://gmod.org/)                     216-392-3087
>> >> Ontario Institute for Cancer Research
>> >
>> >
>>
>>
>>
>> --
>> ------------------------------------------------------------------------
>> Scott Cain, Ph. D.                                   scott at scottcain
>> dot net
>> GMOD Coordinator (http://gmod.org/)                     216-392-3087
>> Ontario Institute for Cancer Research
>
>



--
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     216-392-3087
Ontario Institute for Cancer Research

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Gmod-gbrowse mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse
Reply | Threaded
Open this post in threaded view
|

Re: [Gmod-help] Can I ask a question of Gbrowse2?

Jiantao Yu
Hi Scott,

I have sent the data to your email box.

Jiantao

2012/4/27 Scott Cain <[hidden email]>
Hi Jiantao,

The easiest way to tell what version of GBrowse you are using is the
location of the config files: if they are int /etc/gbrowse, it's
GBrowse2.  If not, it's probably GBrowse 1.

Since part of the problem you are having is related to how your data
are arranged in one track, could you send me a set of real data for
one of those tracks rather than a small sample that don't really
represent the problem?

Thanks,
Scott


On Fri, Apr 27, 2012 at 12:51 PM, Jiantao Yu <[hidden email]> wrote:
> Hi Scott,
>
> I think I am using Gbrowse2, and I made the configure file based on your
> instruction file, Generic Genome Browser Version 2: A Tutorial for
> Administrators.
>
> However, I am not 100 percent sure of this. Can you tell me how can I check
> the version of Gbrowse.
>
> Another point is that, you see, GBrowse2 didn't provide the configure file
> of Arabidopsis, so we generated this file by ourselves. However, the file is
> made by my colleague, so I am not sure if he took this configure file from
> the GBrowse1.x or Gbrowse2.x . If you are sure the configure file is from
> GBrowse1.x, I will ask my colleague and make sure of this. Thanks.
>
>
> Jiantao
>
> 2012/4/27 Scott Cain <[hidden email]>
>>
>> Hi Jiantao,
>>
>> I forgot to ask--what version of GBrowse are you using?  Your conf
>> file looks like GBrowse 1.X.  Also, I trimmed the GMOD help desk email
>> off the cc list.
>>
>> For xyplots: having the boundaries agacent shouldn't matter; that is
>> what the data in the tutorial are, and it works fine.  There is a
>> problem either with your data, your config or both.  I'm working on a
>> simplified case to see if I can reproduce your problem.
>>
>> For the key/track title problem, I'm guessing you had illigal characters.
>>
>> Scott
>>
>>
>> On Fri, Apr 27, 2012 at 12:13 PM, Jiantao Yu <[hidden email]>
>> wrote:
>> > Hi Scott,
>> >
>> > Thanks for reply. I found some answers for the three questions:
>> > i) The reason that the boxes are displayed in hierarchy, I think, would
>> > be
>> > the boxes are too close. For example, if the start and end positions are
>> > like this:
>> >
>> > 1    10    0.8    .    .
>> > 11    20    0.7    .    .
>> > 21    30    0.7    .    .
>> > ...
>> >
>> > They would probably be in hierarchy when I zoom in. However, if I
>> > changed
>> > the positions like this (+ or - 4):
>> >
>> > 4    7    0.8    .    .
>> > 14    17    0.7    .    .
>> > 24    27    0.7    .    .
>> >
>> > They would be on the same line. However, in the case of my data, it is
>> > hard
>> > to determine what values should be added or subtracted. For example, I
>> > used
>> > 30 as the value, it wasn't enough, I mean, 30 is not enough to ensure
>> > all of
>> > the boxes are on the same line for every scale of zooming in or zooming
>> > out.
>> >
>> > From another point, I really hope the width of boxes are still kept
>> > continuous, you see, no gaps.
>> >
>> > ii) I tried to gave all of the entries, located in the same chromosome,
>> > the
>> > same IDs. This method did solve the problem, i.e. the boxes are fixed on
>> > the
>> > same line. However, as you told in the last email, this led to a
>> > problem,
>> > the performance became poor, and the gbrowse couldn't display the curves
>> > sometimes, if there were lots of boxes which had the same IDs.
>> >
>> > So I think the best way would be, can we find a way to display them on
>> > the
>> > same line, while no limitation of their IDs?
>> >
>> > iii) The reason that there is no title of some tracks, is not very
>> > clear.
>> > However, when I changed the 'key' with some other words, the title can
>> > be
>> > normally displayed. So this has not been a problem now.
>> >
>> > Thanks very much for your patience.
>> >
>> > Regards
>> > Jiantao
>> >
>> >
>> >
>> > 2012/4/27 Scott Cain <[hidden email]>
>> >>
>> >> Jiantao,
>> >>
>> >> Please always respond to the mailing list too (reply-all).
>> >>
>> >> No, you absolutely should not try to group these features by ID.  That
>> >> will result in the database loader putting all of the features
>> >> together into one gigantic object in the database, and your
>> >> performance will be very poor (GBrowse may stop rendering those tracks
>> >> altogether).  I'll try to reproduce your problem now.
>> >>
>> >> Scott
>> >>
>> >>
>> >> On Tue, Apr 24, 2012 at 12:23 PM, Jiantao Yu
>> >> <[hidden email]>
>> >> wrote:
>> >> > Hi Scott,
>> >> >
>> >> > I still got the picture like the attahed file, although I changed
>> >> > group_on=subject. Do you think I need to add 'ID=1' for all of the
>> >> > entries
>> >> > in .gff3 file, in order to let them displayed on the single line?
>> >> >
>> >> > GBrowse may group entries according to their IDs? However, when I did
>> >> > this,
>> >> > I can't get the specific balloon for each position/block in the
>> >> > picture,
>> >> > you
>> >> > see, I just got a very general description in the balloon, such as
>> >> > this
>> >> > position is from 1..30Mbp, which is not what I want to get.
>> >> >
>> >> > Thanks a lot.
>> >> >
>> >> >
>> >> > Jiantao
>> >> >
>> >> > 2012/4/24 Scott Cain <[hidden email]>
>> >> >>
>> >> >> Hi Jiantao,
>> >> >>
>> >> >> "subject" should work for your data.  Did you try panning or zooming
>> >> >> to another location to make sure GBrowse wasn't using a cached image
>> >> >> (just reloading isn't enough)?
>> >> >>
>> >> >> Scott
>> >> >>
>> >> >>
>> >> >> On Tue, Apr 24, 2012 at 11:44 AM, Jiantao Yu
>> >> >> <[hidden email]>
>> >> >> wrote:
>> >> >> > Hi Scott,
>> >> >> >
>> >> >> > Really thanks for you quick reply. I tried to change group-on in
>> >> >> > the
>> >> >> > config
>> >> >> > file into
>> >> >> > group_on = display_name                    or
>> >> >> > group_on =subject
>> >> >> >
>> >> >> > However, it seems no effect. I have attached the .gff3 and .conf
>> >> >> > files,
>> >> >> > and
>> >> >> > the configure script for the .gff3 data starts at the line of
>> >> >> > No.1365.
>> >> >> >
>> >> >> > Could you please have a look at them? Thanks very much.
>> >> >> >
>> >> >> > Best regards
>> >> >> > Jiantao
>> >> >> >
>> >> >> >
>> >> >> > 2012/4/24 Scott Cain <[hidden email]>
>> >> >> >>
>> >> >> >> Hello Jiantao,
>> >> >> >>
>> >> >> >> What you are seeing is almost always a result of the config and
>> >> >> >> the
>> >> >> >> data not matching.  I think your "group_on" setting is incorrect;
>> >> >> >> it
>> >> >> >> is usually something like "group_on = display_name" or "group_on
>> >> >> >> =
>> >> >> >> subject".  I don't think setting it equal to the name of the
>> >> >> >> feature
>> >> >> >> type does anything.  If fixing that doesn't work, you could send
>> >> >> >> a
>> >> >> >> sample of the GFF to the list.  Also, there is a section of the
>> >> >> >> tutorial that came with GBrowse that covers xyplots that might
>> >> >> >> help
>> >> >> >> (look for "quantitative data" in the tutorial).
>> >> >> >>
>> >> >> >> Scott
>> >> >> >>
>> >> >> >>
>> >> >> >> On Mon, Apr 23, 2012 at 8:14 PM, Jiantao Yu
>> >> >> >> <[hidden email]>
>> >> >> >> wrote:
>> >> >> >> > Dear Sir/Madam,
>> >> >> >> >
>> >> >> >> > I highly appreciate the smart of GBrowse. When I tried to
>> >> >> >> > display
>> >> >> >> > the
>> >> >> >> > quantitative data, and set up the zoom with small values, such
>> >> >> >> > as
>> >> >> >> > 40Kbp,
>> >> >> >> > I
>> >> >> >> > got the picture as is attached (picture1.png). However, I
>> >> >> >> > perfer
>> >> >> >> > the
>> >> >> >> > blocks
>> >> >> >> > to locate on the same line, rather than being hierarchical. I
>> >> >> >> > mean,
>> >> >> >> > It
>> >> >> >> > would
>> >> >> >> > be better if the blocks are kept as they are in the zoom of
>> >> >> >> > 500Kbp
>> >> >> >> > (all
>> >> >> >> > of
>> >> >> >> > the blocks are at one line, shown in the picture2.png). Could
>> >> >> >> > you
>> >> >> >> > please
>> >> >> >> > tell me how to do this? Many thanks!
>> >> >> >> >
>> >> >> >> > BTW: my configure script is like this:
>> >> >> >> >
>> >> >> >> > [methySW_MSH1_1000K]
>> >> >> >> > feature         = methySW_MSH1_1000K
>> >> >> >> > glyph           = xyplot
>> >> >> >> > graph_type      = boxes
>> >> >> >> > fgcolor         = red
>> >> >> >> > bgcolor         = red
>> >> >> >> > height          = 80
>> >> >> >> > min_score       = 0
>> >> >> >> > max_score       = 1
>> >> >> >> > scale           = none
>> >> >> >> > group_on        = methySW_MSH1_1000K
>> >> >> >> > category        = Quantitative Data
>> >> >> >> > label           = 0
>> >> >> >> > key             = Methylation Profile - MSH1_dr_vs_Col0
>> >> >> >> > (Sliding
>> >> >> >> > Windows:1000K-size)
>> >> >> >> >
>> >> >> >> >
>> >> >> >> > Best wishes,
>> >> >> >> > Jiantao
>> >> >> >>
>> >> >> >>
>> >> >> >>
>> >> >> >> --
>> >> >> >>
>> >> >> >>
>> >> >> >>
>> >> >> >> ------------------------------------------------------------------------
>> >> >> >> Scott Cain, Ph. D.                                   scott at
>> >> >> >> scottcain
>> >> >> >> dot net
>> >> >> >> GMOD Coordinator (http://gmod.org/)
>> >> >> >> <a href="tel:216-392-3087" value="+12163923087">216-392-3087
>> >> >> >> Ontario Institute for Cancer Research
>> >> >> >
>> >> >> >
>> >> >>
>> >> >>
>> >> >>
>> >> >> --
>> >> >>
>> >> >>
>> >> >> ------------------------------------------------------------------------
>> >> >> Scott Cain, Ph. D.                                   scott at
>> >> >> scottcain
>> >> >> dot net
>> >> >> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087">216-392-3087
>> >> >> Ontario Institute for Cancer Research
>> >> >
>> >> >
>> >>
>> >>
>> >>
>> >> --
>> >>
>> >> ------------------------------------------------------------------------
>> >> Scott Cain, Ph. D.                                   scott at scottcain
>> >> dot net
>> >> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087">216-392-3087
>> >> Ontario Institute for Cancer Research
>> >
>> >
>>
>>
>>
>> --
>> ------------------------------------------------------------------------
>> Scott Cain, Ph. D.                                   scott at scottcain
>> dot net
>> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087">216-392-3087
>> Ontario Institute for Cancer Research
>
>



--
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087">216-392-3087
Ontario Institute for Cancer Research


------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Gmod-gbrowse mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse
Reply | Threaded
Open this post in threaded view
|

Re: [Gmod-help] Can I ask a question of Gbrowse2?

Scott Cain
In reply to this post by Jiantao Yu
Hi Jiantao,

I managed to reproduce your problem.  It turns out that "group on =
source" doesn't appear to work with the current bioperl.  I fixed it
for the sample data you gave me (for Chr1 only, and one source only),
by adding to each line a "Name=fake;" and then changing the config to
do "group on = display_name".  So, for example, one line of the GFF
would look like this:

Chr1    MSH1_dr_vs_Col_0   meylaSW_MSH1_1000K_feature   475006  524995
 0.777166    .   .   Name=fake;Note=Wi..

Scott


On Fri, Apr 27, 2012 at 12:51 PM, Jiantao Yu <[hidden email]> wrote:

> Hi Scott,
>
> I think I am using Gbrowse2, and I made the configure file based on your
> instruction file, Generic Genome Browser Version 2: A Tutorial for
> Administrators.
>
> However, I am not 100 percent sure of this. Can you tell me how can I check
> the version of Gbrowse.
>
> Another point is that, you see, GBrowse2 didn't provide the configure file
> of Arabidopsis, so we generated this file by ourselves. However, the file is
> made by my colleague, so I am not sure if he took this configure file from
> the GBrowse1.x or Gbrowse2.x . If you are sure the configure file is from
> GBrowse1.x, I will ask my colleague and make sure of this. Thanks.
>
>
> Jiantao
>
> 2012/4/27 Scott Cain <[hidden email]>
>>
>> Hi Jiantao,
>>
>> I forgot to ask--what version of GBrowse are you using?  Your conf
>> file looks like GBrowse 1.X.  Also, I trimmed the GMOD help desk email
>> off the cc list.
>>
>> For xyplots: having the boundaries agacent shouldn't matter; that is
>> what the data in the tutorial are, and it works fine.  There is a
>> problem either with your data, your config or both.  I'm working on a
>> simplified case to see if I can reproduce your problem.
>>
>> For the key/track title problem, I'm guessing you had illigal characters.
>>
>> Scott
>>
>>
>> On Fri, Apr 27, 2012 at 12:13 PM, Jiantao Yu <[hidden email]>
>> wrote:
>> > Hi Scott,
>> >
>> > Thanks for reply. I found some answers for the three questions:
>> > i) The reason that the boxes are displayed in hierarchy, I think, would
>> > be
>> > the boxes are too close. For example, if the start and end positions are
>> > like this:
>> >
>> > 1    10    0.8    .    .
>> > 11    20    0.7    .    .
>> > 21    30    0.7    .    .
>> > ...
>> >
>> > They would probably be in hierarchy when I zoom in. However, if I
>> > changed
>> > the positions like this (+ or - 4):
>> >
>> > 4    7    0.8    .    .
>> > 14    17    0.7    .    .
>> > 24    27    0.7    .    .
>> >
>> > They would be on the same line. However, in the case of my data, it is
>> > hard
>> > to determine what values should be added or subtracted. For example, I
>> > used
>> > 30 as the value, it wasn't enough, I mean, 30 is not enough to ensure
>> > all of
>> > the boxes are on the same line for every scale of zooming in or zooming
>> > out.
>> >
>> > From another point, I really hope the width of boxes are still kept
>> > continuous, you see, no gaps.
>> >
>> > ii) I tried to gave all of the entries, located in the same chromosome,
>> > the
>> > same IDs. This method did solve the problem, i.e. the boxes are fixed on
>> > the
>> > same line. However, as you told in the last email, this led to a
>> > problem,
>> > the performance became poor, and the gbrowse couldn't display the curves
>> > sometimes, if there were lots of boxes which had the same IDs.
>> >
>> > So I think the best way would be, can we find a way to display them on
>> > the
>> > same line, while no limitation of their IDs?
>> >
>> > iii) The reason that there is no title of some tracks, is not very
>> > clear.
>> > However, when I changed the 'key' with some other words, the title can
>> > be
>> > normally displayed. So this has not been a problem now.
>> >
>> > Thanks very much for your patience.
>> >
>> > Regards
>> > Jiantao
>> >
>> >
>> >
>> > 2012/4/27 Scott Cain <[hidden email]>
>> >>
>> >> Jiantao,
>> >>
>> >> Please always respond to the mailing list too (reply-all).
>> >>
>> >> No, you absolutely should not try to group these features by ID.  That
>> >> will result in the database loader putting all of the features
>> >> together into one gigantic object in the database, and your
>> >> performance will be very poor (GBrowse may stop rendering those tracks
>> >> altogether).  I'll try to reproduce your problem now.
>> >>
>> >> Scott
>> >>
>> >>
>> >> On Tue, Apr 24, 2012 at 12:23 PM, Jiantao Yu
>> >> <[hidden email]>
>> >> wrote:
>> >> > Hi Scott,
>> >> >
>> >> > I still got the picture like the attahed file, although I changed
>> >> > group_on=subject. Do you think I need to add 'ID=1' for all of the
>> >> > entries
>> >> > in .gff3 file, in order to let them displayed on the single line?
>> >> >
>> >> > GBrowse may group entries according to their IDs? However, when I did
>> >> > this,
>> >> > I can't get the specific balloon for each position/block in the
>> >> > picture,
>> >> > you
>> >> > see, I just got a very general description in the balloon, such as
>> >> > this
>> >> > position is from 1..30Mbp, which is not what I want to get.
>> >> >
>> >> > Thanks a lot.
>> >> >
>> >> >
>> >> > Jiantao
>> >> >
>> >> > 2012/4/24 Scott Cain <[hidden email]>
>> >> >>
>> >> >> Hi Jiantao,
>> >> >>
>> >> >> "subject" should work for your data.  Did you try panning or zooming
>> >> >> to another location to make sure GBrowse wasn't using a cached image
>> >> >> (just reloading isn't enough)?
>> >> >>
>> >> >> Scott
>> >> >>
>> >> >>
>> >> >> On Tue, Apr 24, 2012 at 11:44 AM, Jiantao Yu
>> >> >> <[hidden email]>
>> >> >> wrote:
>> >> >> > Hi Scott,
>> >> >> >
>> >> >> > Really thanks for you quick reply. I tried to change group-on in
>> >> >> > the
>> >> >> > config
>> >> >> > file into
>> >> >> > group_on = display_name                    or
>> >> >> > group_on =subject
>> >> >> >
>> >> >> > However, it seems no effect. I have attached the .gff3 and .conf
>> >> >> > files,
>> >> >> > and
>> >> >> > the configure script for the .gff3 data starts at the line of
>> >> >> > No.1365.
>> >> >> >
>> >> >> > Could you please have a look at them? Thanks very much.
>> >> >> >
>> >> >> > Best regards
>> >> >> > Jiantao
>> >> >> >
>> >> >> >
>> >> >> > 2012/4/24 Scott Cain <[hidden email]>
>> >> >> >>
>> >> >> >> Hello Jiantao,
>> >> >> >>
>> >> >> >> What you are seeing is almost always a result of the config and
>> >> >> >> the
>> >> >> >> data not matching.  I think your "group_on" setting is incorrect;
>> >> >> >> it
>> >> >> >> is usually something like "group_on = display_name" or "group_on
>> >> >> >> =
>> >> >> >> subject".  I don't think setting it equal to the name of the
>> >> >> >> feature
>> >> >> >> type does anything.  If fixing that doesn't work, you could send
>> >> >> >> a
>> >> >> >> sample of the GFF to the list.  Also, there is a section of the
>> >> >> >> tutorial that came with GBrowse that covers xyplots that might
>> >> >> >> help
>> >> >> >> (look for "quantitative data" in the tutorial).
>> >> >> >>
>> >> >> >> Scott
>> >> >> >>
>> >> >> >>
>> >> >> >> On Mon, Apr 23, 2012 at 8:14 PM, Jiantao Yu
>> >> >> >> <[hidden email]>
>> >> >> >> wrote:
>> >> >> >> > Dear Sir/Madam,
>> >> >> >> >
>> >> >> >> > I highly appreciate the smart of GBrowse. When I tried to
>> >> >> >> > display
>> >> >> >> > the
>> >> >> >> > quantitative data, and set up the zoom with small values, such
>> >> >> >> > as
>> >> >> >> > 40Kbp,
>> >> >> >> > I
>> >> >> >> > got the picture as is attached (picture1.png). However, I
>> >> >> >> > perfer
>> >> >> >> > the
>> >> >> >> > blocks
>> >> >> >> > to locate on the same line, rather than being hierarchical. I
>> >> >> >> > mean,
>> >> >> >> > It
>> >> >> >> > would
>> >> >> >> > be better if the blocks are kept as they are in the zoom of
>> >> >> >> > 500Kbp
>> >> >> >> > (all
>> >> >> >> > of
>> >> >> >> > the blocks are at one line, shown in the picture2.png). Could
>> >> >> >> > you
>> >> >> >> > please
>> >> >> >> > tell me how to do this? Many thanks!
>> >> >> >> >
>> >> >> >> > BTW: my configure script is like this:
>> >> >> >> >
>> >> >> >> > [methySW_MSH1_1000K]
>> >> >> >> > feature         = methySW_MSH1_1000K
>> >> >> >> > glyph           = xyplot
>> >> >> >> > graph_type      = boxes
>> >> >> >> > fgcolor         = red
>> >> >> >> > bgcolor         = red
>> >> >> >> > height          = 80
>> >> >> >> > min_score       = 0
>> >> >> >> > max_score       = 1
>> >> >> >> > scale           = none
>> >> >> >> > group_on        = methySW_MSH1_1000K
>> >> >> >> > category        = Quantitative Data
>> >> >> >> > label           = 0
>> >> >> >> > key             = Methylation Profile - MSH1_dr_vs_Col0
>> >> >> >> > (Sliding
>> >> >> >> > Windows:1000K-size)
>> >> >> >> >
>> >> >> >> >
>> >> >> >> > Best wishes,
>> >> >> >> > Jiantao
>> >> >> >>
>> >> >> >>
>> >> >> >>
>> >> >> >> --
>> >> >> >>
>> >> >> >>
>> >> >> >>
>> >> >> >> ------------------------------------------------------------------------
>> >> >> >> Scott Cain, Ph. D.                                   scott at
>> >> >> >> scottcain
>> >> >> >> dot net
>> >> >> >> GMOD Coordinator (http://gmod.org/)
>> >> >> >> 216-392-3087
>> >> >> >> Ontario Institute for Cancer Research
>> >> >> >
>> >> >> >
>> >> >>
>> >> >>
>> >> >>
>> >> >> --
>> >> >>
>> >> >>
>> >> >> ------------------------------------------------------------------------
>> >> >> Scott Cain, Ph. D.                                   scott at
>> >> >> scottcain
>> >> >> dot net
>> >> >> GMOD Coordinator (http://gmod.org/)                     216-392-3087
>> >> >> Ontario Institute for Cancer Research
>> >> >
>> >> >
>> >>
>> >>
>> >>
>> >> --
>> >>
>> >> ------------------------------------------------------------------------
>> >> Scott Cain, Ph. D.                                   scott at scottcain
>> >> dot net
>> >> GMOD Coordinator (http://gmod.org/)                     216-392-3087
>> >> Ontario Institute for Cancer Research
>> >
>> >
>>
>>
>>
>> --
>> ------------------------------------------------------------------------
>> Scott Cain, Ph. D.                                   scott at scottcain
>> dot net
>> GMOD Coordinator (http://gmod.org/)                     216-392-3087
>> Ontario Institute for Cancer Research
>
>



--
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     216-392-3087
Ontario Institute for Cancer Research
Hi

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Gmod-gbrowse mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse
Reply | Threaded
Open this post in threaded view
|

Re: [Gmod-help] Can I ask a question of Gbrowse2?

Jiantao Yu
Hi Scott,

Many thanks for your help! It has been successful!! Could you please tell me the reason why we can give the same name to many entries, however, shouldn't give the same IDs to many entries. Except the reason you said that a big bulk of data will belong to one feature in this case, is there any other reason for not doing this? I am not sure why the performance will become poor?

By the way, in the case of my data, I want to group one kind of name, fake, into subtrack, and group another kind of name, fake2, into another subtrack, and give them different colors to display. I used this script, but it looked useless. Could you please help me again? thanks for your patience.

[solveScatter3]
feature         = scatter_feature3
glyph           = xyplot
graph_type      = boxes
subtrack select = Strand strand
subtrack table  = :OverMethy +1 ;
                              :UnderMethy -1
subtrack select fgcolor = sub {
                    my $f = shift;
                    my ($name) = $f->attributes('Name');
                    return "hotpink" if($name =~ /fake/);
                    return "blue" if($name =~ /fake1/);
                }
subtrack select bgcolor = sub {
                    my $f = shift;
                    my ($name) = $f->attributes('Name');
                    return "hotpink" if($name =~ /fake/);
                    return "blue" if($name =~ /fake1/);
                }
label        = 0
height          = 80
min_score       = 0
max_score       = 205
scale           = none
group_on        = display_name
category        = Quantitative Data
key             = solveScatter3

The last question is, why can 'scale' NOT be set up as 'left' or 'right', or else, there are many scale-coordinates (y-axis) appearing in the screen?

Thanks again.
 
Jiantao

2012/4/27 Scott Cain <[hidden email]>
Hi Jiantao,

I managed to reproduce your problem.  It turns out that "group on =
source" doesn't appear to work with the current bioperl.  I fixed it
for the sample data you gave me (for Chr1 only, and one source only),
by adding to each line a "Name=fake;" and then changing the config to
do "group on = display_name".  So, for example, one line of the GFF
would look like this:

Chr1    MSH1_dr_vs_Col_0   meylaSW_MSH1_1000K_feature   475006  524995
 0.777166    .   .   Name=fake;Note=Wi..

Scott


On Fri, Apr 27, 2012 at 12:51 PM, Jiantao Yu <[hidden email]> wrote:
> Hi Scott,
>
> I think I am using Gbrowse2, and I made the configure file based on your
> instruction file, Generic Genome Browser Version 2: A Tutorial for
> Administrators.
>
> However, I am not 100 percent sure of this. Can you tell me how can I check
> the version of Gbrowse.
>
> Another point is that, you see, GBrowse2 didn't provide the configure file
> of Arabidopsis, so we generated this file by ourselves. However, the file is
> made by my colleague, so I am not sure if he took this configure file from
> the GBrowse1.x or Gbrowse2.x . If you are sure the configure file is from
> GBrowse1.x, I will ask my colleague and make sure of this. Thanks.
>
>
> Jiantao
>
> 2012/4/27 Scott Cain <[hidden email]>
>>
>> Hi Jiantao,
>>
>> I forgot to ask--what version of GBrowse are you using?  Your conf
>> file looks like GBrowse 1.X.  Also, I trimmed the GMOD help desk email
>> off the cc list.
>>
>> For xyplots: having the boundaries agacent shouldn't matter; that is
>> what the data in the tutorial are, and it works fine.  There is a
>> problem either with your data, your config or both.  I'm working on a
>> simplified case to see if I can reproduce your problem.
>>
>> For the key/track title problem, I'm guessing you had illigal characters.
>>
>> Scott
>>
>>
>> On Fri, Apr 27, 2012 at 12:13 PM, Jiantao Yu <[hidden email]>
>> wrote:
>> > Hi Scott,
>> >
>> > Thanks for reply. I found some answers for the three questions:
>> > i) The reason that the boxes are displayed in hierarchy, I think, would
>> > be
>> > the boxes are too close. For example, if the start and end positions are
>> > like this:
>> >
>> > 1    10    0.8    .    .
>> > 11    20    0.7    .    .
>> > 21    30    0.7    .    .
>> > ...
>> >
>> > They would probably be in hierarchy when I zoom in. However, if I
>> > changed
>> > the positions like this (+ or - 4):
>> >
>> > 4    7    0.8    .    .
>> > 14    17    0.7    .    .
>> > 24    27    0.7    .    .
>> >
>> > They would be on the same line. However, in the case of my data, it is
>> > hard
>> > to determine what values should be added or subtracted. For example, I
>> > used
>> > 30 as the value, it wasn't enough, I mean, 30 is not enough to ensure
>> > all of
>> > the boxes are on the same line for every scale of zooming in or zooming
>> > out.
>> >
>> > From another point, I really hope the width of boxes are still kept
>> > continuous, you see, no gaps.
>> >
>> > ii) I tried to gave all of the entries, located in the same chromosome,
>> > the
>> > same IDs. This method did solve the problem, i.e. the boxes are fixed on
>> > the
>> > same line. However, as you told in the last email, this led to a
>> > problem,
>> > the performance became poor, and the gbrowse couldn't display the curves
>> > sometimes, if there were lots of boxes which had the same IDs.
>> >
>> > So I think the best way would be, can we find a way to display them on
>> > the
>> > same line, while no limitation of their IDs?
>> >
>> > iii) The reason that there is no title of some tracks, is not very
>> > clear.
>> > However, when I changed the 'key' with some other words, the title can
>> > be
>> > normally displayed. So this has not been a problem now.
>> >
>> > Thanks very much for your patience.
>> >
>> > Regards
>> > Jiantao
>> >
>> >
>> >
>> > 2012/4/27 Scott Cain <[hidden email]>
>> >>
>> >> Jiantao,
>> >>
>> >> Please always respond to the mailing list too (reply-all).
>> >>
>> >> No, you absolutely should not try to group these features by ID.  That
>> >> will result in the database loader putting all of the features
>> >> together into one gigantic object in the database, and your
>> >> performance will be very poor (GBrowse may stop rendering those tracks
>> >> altogether).  I'll try to reproduce your problem now.
>> >>
>> >> Scott
>> >>
>> >>
>> >> On Tue, Apr 24, 2012 at 12:23 PM, Jiantao Yu
>> >> <[hidden email]>
>> >> wrote:
>> >> > Hi Scott,
>> >> >
>> >> > I still got the picture like the attahed file, although I changed
>> >> > group_on=subject. Do you think I need to add 'ID=1' for all of the
>> >> > entries
>> >> > in .gff3 file, in order to let them displayed on the single line?
>> >> >
>> >> > GBrowse may group entries according to their IDs? However, when I did
>> >> > this,
>> >> > I can't get the specific balloon for each position/block in the
>> >> > picture,
>> >> > you
>> >> > see, I just got a very general description in the balloon, such as
>> >> > this
>> >> > position is from 1..30Mbp, which is not what I want to get.
>> >> >
>> >> > Thanks a lot.
>> >> >
>> >> >
>> >> > Jiantao
>> >> >
>> >> > 2012/4/24 Scott Cain <[hidden email]>
>> >> >>
>> >> >> Hi Jiantao,
>> >> >>
>> >> >> "subject" should work for your data.  Did you try panning or zooming
>> >> >> to another location to make sure GBrowse wasn't using a cached image
>> >> >> (just reloading isn't enough)?
>> >> >>
>> >> >> Scott
>> >> >>
>> >> >>
>> >> >> On Tue, Apr 24, 2012 at 11:44 AM, Jiantao Yu
>> >> >> <[hidden email]>
>> >> >> wrote:
>> >> >> > Hi Scott,
>> >> >> >
>> >> >> > Really thanks for you quick reply. I tried to change group-on in
>> >> >> > the
>> >> >> > config
>> >> >> > file into
>> >> >> > group_on = display_name                    or
>> >> >> > group_on =subject
>> >> >> >
>> >> >> > However, it seems no effect. I have attached the .gff3 and .conf
>> >> >> > files,
>> >> >> > and
>> >> >> > the configure script for the .gff3 data starts at the line of
>> >> >> > No.1365.
>> >> >> >
>> >> >> > Could you please have a look at them? Thanks very much.
>> >> >> >
>> >> >> > Best regards
>> >> >> > Jiantao
>> >> >> >
>> >> >> >
>> >> >> > 2012/4/24 Scott Cain <[hidden email]>
>> >> >> >>
>> >> >> >> Hello Jiantao,
>> >> >> >>
>> >> >> >> What you are seeing is almost always a result of the config and
>> >> >> >> the
>> >> >> >> data not matching.  I think your "group_on" setting is incorrect;
>> >> >> >> it
>> >> >> >> is usually something like "group_on = display_name" or "group_on
>> >> >> >> =
>> >> >> >> subject".  I don't think setting it equal to the name of the
>> >> >> >> feature
>> >> >> >> type does anything.  If fixing that doesn't work, you could send
>> >> >> >> a
>> >> >> >> sample of the GFF to the list.  Also, there is a section of the
>> >> >> >> tutorial that came with GBrowse that covers xyplots that might
>> >> >> >> help
>> >> >> >> (look for "quantitative data" in the tutorial).
>> >> >> >>
>> >> >> >> Scott
>> >> >> >>
>> >> >> >>
>> >> >> >> On Mon, Apr 23, 2012 at 8:14 PM, Jiantao Yu
>> >> >> >> <[hidden email]>
>> >> >> >> wrote:
>> >> >> >> > Dear Sir/Madam,
>> >> >> >> >
>> >> >> >> > I highly appreciate the smart of GBrowse. When I tried to
>> >> >> >> > display
>> >> >> >> > the
>> >> >> >> > quantitative data, and set up the zoom with small values, such
>> >> >> >> > as
>> >> >> >> > 40Kbp,
>> >> >> >> > I
>> >> >> >> > got the picture as is attached (picture1.png). However, I
>> >> >> >> > perfer
>> >> >> >> > the
>> >> >> >> > blocks
>> >> >> >> > to locate on the same line, rather than being hierarchical. I
>> >> >> >> > mean,
>> >> >> >> > It
>> >> >> >> > would
>> >> >> >> > be better if the blocks are kept as they are in the zoom of
>> >> >> >> > 500Kbp
>> >> >> >> > (all
>> >> >> >> > of
>> >> >> >> > the blocks are at one line, shown in the picture2.png). Could
>> >> >> >> > you
>> >> >> >> > please
>> >> >> >> > tell me how to do this? Many thanks!
>> >> >> >> >
>> >> >> >> > BTW: my configure script is like this:
>> >> >> >> >
>> >> >> >> > [methySW_MSH1_1000K]
>> >> >> >> > feature         = methySW_MSH1_1000K
>> >> >> >> > glyph           = xyplot
>> >> >> >> > graph_type      = boxes
>> >> >> >> > fgcolor         = red
>> >> >> >> > bgcolor         = red
>> >> >> >> > height          = 80
>> >> >> >> > min_score       = 0
>> >> >> >> > max_score       = 1
>> >> >> >> > scale           = none
>> >> >> >> > group_on        = methySW_MSH1_1000K
>> >> >> >> > category        = Quantitative Data
>> >> >> >> > label           = 0
>> >> >> >> > key             = Methylation Profile - MSH1_dr_vs_Col0
>> >> >> >> > (Sliding
>> >> >> >> > Windows:1000K-size)
>> >> >> >> >
>> >> >> >> >
>> >> >> >> > Best wishes,
>> >> >> >> > Jiantao
>> >> >> >>
>> >> >> >>
>> >> >> >>
>> >> >> >> --
>> >> >> >>
>> >> >> >>
>> >> >> >>
>> >> >> >> ------------------------------------------------------------------------
>> >> >> >> Scott Cain, Ph. D.                                   scott at
>> >> >> >> scottcain
>> >> >> >> dot net
>> >> >> >> GMOD Coordinator (http://gmod.org/)
>> >> >> >> <a href="tel:216-392-3087" value="+12163923087">216-392-3087
>> >> >> >> Ontario Institute for Cancer Research
>> >> >> >
>> >> >> >
>> >> >>
>> >> >>
>> >> >>
>> >> >> --
>> >> >>
>> >> >>
>> >> >> ------------------------------------------------------------------------
>> >> >> Scott Cain, Ph. D.                                   scott at
>> >> >> scottcain
>> >> >> dot net
>> >> >> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087">216-392-3087
>> >> >> Ontario Institute for Cancer Research
>> >> >
>> >> >
>> >>
>> >>
>> >>
>> >> --
>> >>
>> >> ------------------------------------------------------------------------
>> >> Scott Cain, Ph. D.                                   scott at scottcain
>> >> dot net
>> >> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087">216-392-3087
>> >> Ontario Institute for Cancer Research
>> >
>> >
>>
>>
>>
>> --
>> ------------------------------------------------------------------------
>> Scott Cain, Ph. D.                                   scott at scottcain
>> dot net
>> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087">216-392-3087
>> Ontario Institute for Cancer Research
>
>



--
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087">216-392-3087
Ontario Institute for Cancer Research
Hi


------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Gmod-gbrowse mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse
Reply | Threaded
Open this post in threaded view
|

Re: [Gmod-help] Can I ask a question of Gbrowse2?

Jiantao Yu-2
Hi Scott

I got the error as following, when I tried to install GBrowser via:

Could not find apache executable on this system. Can't figure out version number for config file. at /home/jiantao/GBrowse/install_util/GBrowseInstall.pm line 932.


I tried to solve this problem by
http://sourceforge.net/p/gmod/mailman/message/29595042/

but it didn't help. Can you help me? Thanks!

Jiantao

2012-04-28 7:53 GMT+08:00 Jiantao Yu <[hidden email]>:
Hi Scott,

Many thanks for your help! It has been successful!! Could you please tell me the reason why we can give the same name to many entries, however, shouldn't give the same IDs to many entries. Except the reason you said that a big bulk of data will belong to one feature in this case, is there any other reason for not doing this? I am not sure why the performance will become poor?

By the way, in the case of my data, I want to group one kind of name, fake, into subtrack, and group another kind of name, fake2, into another subtrack, and give them different colors to display. I used this script, but it looked useless. Could you please help me again? thanks for your patience.

[solveScatter3]
feature         = scatter_feature3
glyph           = xyplot
graph_type      = boxes
subtrack select = Strand strand
subtrack table  = :OverMethy +1 ;
                              :UnderMethy -1
subtrack select fgcolor = sub {
                    my $f = shift;
                    my ($name) = $f->attributes('Name');
                    return "hotpink" if($name =~ /fake/);
                    return "blue" if($name =~ /fake1/);
                }
subtrack select bgcolor = sub {
                    my $f = shift;
                    my ($name) = $f->attributes('Name');
                    return "hotpink" if($name =~ /fake/);
                    return "blue" if($name =~ /fake1/);
                }
label        = 0
height          = 80
min_score       = 0
max_score       = 205
scale           = none
group_on        = display_name
category        = Quantitative Data
key             = solveScatter3

The last question is, why can 'scale' NOT be set up as 'left' or 'right', or else, there are many scale-coordinates (y-axis) appearing in the screen?

Thanks again.

 
Jiantao

2012/4/27 Scott Cain <[hidden email]>
Hi Jiantao,

I managed to reproduce your problem.  It turns out that "group on =
source" doesn't appear to work with the current bioperl.  I fixed it
for the sample data you gave me (for Chr1 only, and one source only),
by adding to each line a "Name=fake;" and then changing the config to
do "group on = display_name".  So, for example, one line of the GFF
would look like this:

Chr1    MSH1_dr_vs_Col_0   meylaSW_MSH1_1000K_feature   475006  524995
 0.777166    .   .   Name=fake;Note=Wi..

Scott


On Fri, Apr 27, 2012 at 12:51 PM, Jiantao Yu <[hidden email]> wrote:
> Hi Scott,
>
> I think I am using Gbrowse2, and I made the configure file based on your
> instruction file, Generic Genome Browser Version 2: A Tutorial for
> Administrators.
>
> However, I am not 100 percent sure of this. Can you tell me how can I check
> the version of Gbrowse.
>
> Another point is that, you see, GBrowse2 didn't provide the configure file
> of Arabidopsis, so we generated this file by ourselves. However, the file is
> made by my colleague, so I am not sure if he took this configure file from
> the GBrowse1.x or Gbrowse2.x . If you are sure the configure file is from
> GBrowse1.x, I will ask my colleague and make sure of this. Thanks.
>
>
> Jiantao
>
> 2012/4/27 Scott Cain <[hidden email]>
>>
>> Hi Jiantao,
>>
>> I forgot to ask--what version of GBrowse are you using?  Your conf
>> file looks like GBrowse 1.X.  Also, I trimmed the GMOD help desk email
>> off the cc list.
>>
>> For xyplots: having the boundaries agacent shouldn't matter; that is
>> what the data in the tutorial are, and it works fine.  There is a
>> problem either with your data, your config or both.  I'm working on a
>> simplified case to see if I can reproduce your problem.
>>
>> For the key/track title problem, I'm guessing you had illigal characters.
>>
>> Scott
>>
>>
>> On Fri, Apr 27, 2012 at 12:13 PM, Jiantao Yu <[hidden email]>
>> wrote:
>> > Hi Scott,
>> >
>> > Thanks for reply. I found some answers for the three questions:
>> > i) The reason that the boxes are displayed in hierarchy, I think, would
>> > be
>> > the boxes are too close. For example, if the start and end positions are
>> > like this:
>> >
>> > 1    10    0.8    .    .
>> > 11    20    0.7    .    .
>> > 21    30    0.7    .    .
>> > ...
>> >
>> > They would probably be in hierarchy when I zoom in. However, if I
>> > changed
>> > the positions like this (+ or - 4):
>> >
>> > 4    7    0.8    .    .
>> > 14    17    0.7    .    .
>> > 24    27    0.7    .    .
>> >
>> > They would be on the same line. However, in the case of my data, it is
>> > hard
>> > to determine what values should be added or subtracted. For example, I
>> > used
>> > 30 as the value, it wasn't enough, I mean, 30 is not enough to ensure
>> > all of
>> > the boxes are on the same line for every scale of zooming in or zooming
>> > out.
>> >
>> > From another point, I really hope the width of boxes are still kept
>> > continuous, you see, no gaps.
>> >
>> > ii) I tried to gave all of the entries, located in the same chromosome,
>> > the
>> > same IDs. This method did solve the problem, i.e. the boxes are fixed on
>> > the
>> > same line. However, as you told in the last email, this led to a
>> > problem,
>> > the performance became poor, and the gbrowse couldn't display the curves
>> > sometimes, if there were lots of boxes which had the same IDs.
>> >
>> > So I think the best way would be, can we find a way to display them on
>> > the
>> > same line, while no limitation of their IDs?
>> >
>> > iii) The reason that there is no title of some tracks, is not very
>> > clear.
>> > However, when I changed the 'key' with some other words, the title can
>> > be
>> > normally displayed. So this has not been a problem now.
>> >
>> > Thanks very much for your patience.
>> >
>> > Regards
>> > Jiantao
>> >
>> >
>> >
>> > 2012/4/27 Scott Cain <[hidden email]>
>> >>
>> >> Jiantao,
>> >>
>> >> Please always respond to the mailing list too (reply-all).
>> >>
>> >> No, you absolutely should not try to group these features by ID.  That
>> >> will result in the database loader putting all of the features
>> >> together into one gigantic object in the database, and your
>> >> performance will be very poor (GBrowse may stop rendering those tracks
>> >> altogether).  I'll try to reproduce your problem now.
>> >>
>> >> Scott
>> >>
>> >>
>> >> On Tue, Apr 24, 2012 at 12:23 PM, Jiantao Yu
>> >> <[hidden email]>
>> >> wrote:
>> >> > Hi Scott,
>> >> >
>> >> > I still got the picture like the attahed file, although I changed
>> >> > group_on=subject. Do you think I need to add 'ID=1' for all of the
>> >> > entries
>> >> > in .gff3 file, in order to let them displayed on the single line?
>> >> >
>> >> > GBrowse may group entries according to their IDs? However, when I did
>> >> > this,
>> >> > I can't get the specific balloon for each position/block in the
>> >> > picture,
>> >> > you
>> >> > see, I just got a very general description in the balloon, such as
>> >> > this
>> >> > position is from 1..30Mbp, which is not what I want to get.
>> >> >
>> >> > Thanks a lot.
>> >> >
>> >> >
>> >> > Jiantao
>> >> >
>> >> > 2012/4/24 Scott Cain <[hidden email]>
>> >> >>
>> >> >> Hi Jiantao,
>> >> >>
>> >> >> "subject" should work for your data.  Did you try panning or zooming
>> >> >> to another location to make sure GBrowse wasn't using a cached image
>> >> >> (just reloading isn't enough)?
>> >> >>
>> >> >> Scott
>> >> >>
>> >> >>
>> >> >> On Tue, Apr 24, 2012 at 11:44 AM, Jiantao Yu
>> >> >> <[hidden email]>
>> >> >> wrote:
>> >> >> > Hi Scott,
>> >> >> >
>> >> >> > Really thanks for you quick reply. I tried to change group-on in
>> >> >> > the
>> >> >> > config
>> >> >> > file into
>> >> >> > group_on = display_name                    or
>> >> >> > group_on =subject
>> >> >> >
>> >> >> > However, it seems no effect. I have attached the .gff3 and .conf
>> >> >> > files,
>> >> >> > and
>> >> >> > the configure script for the .gff3 data starts at the line of
>> >> >> > No.1365.
>> >> >> >
>> >> >> > Could you please have a look at them? Thanks very much.
>> >> >> >
>> >> >> > Best regards
>> >> >> > Jiantao
>> >> >> >
>> >> >> >
>> >> >> > 2012/4/24 Scott Cain <[hidden email]>
>> >> >> >>
>> >> >> >> Hello Jiantao,
>> >> >> >>
>> >> >> >> What you are seeing is almost always a result of the config and
>> >> >> >> the
>> >> >> >> data not matching.  I think your "group_on" setting is incorrect;
>> >> >> >> it
>> >> >> >> is usually something like "group_on = display_name" or "group_on
>> >> >> >> =
>> >> >> >> subject".  I don't think setting it equal to the name of the
>> >> >> >> feature
>> >> >> >> type does anything.  If fixing that doesn't work, you could send
>> >> >> >> a
>> >> >> >> sample of the GFF to the list.  Also, there is a section of the
>> >> >> >> tutorial that came with GBrowse that covers xyplots that might
>> >> >> >> help
>> >> >> >> (look for "quantitative data" in the tutorial).
>> >> >> >>
>> >> >> >> Scott
>> >> >> >>
>> >> >> >>
>> >> >> >> On Mon, Apr 23, 2012 at 8:14 PM, Jiantao Yu
>> >> >> >> <[hidden email]>
>> >> >> >> wrote:
>> >> >> >> > Dear Sir/Madam,
>> >> >> >> >
>> >> >> >> > I highly appreciate the smart of GBrowse. When I tried to
>> >> >> >> > display
>> >> >> >> > the
>> >> >> >> > quantitative data, and set up the zoom with small values, such
>> >> >> >> > as
>> >> >> >> > 40Kbp,
>> >> >> >> > I
>> >> >> >> > got the picture as is attached (picture1.png). However, I
>> >> >> >> > perfer
>> >> >> >> > the
>> >> >> >> > blocks
>> >> >> >> > to locate on the same line, rather than being hierarchical. I
>> >> >> >> > mean,
>> >> >> >> > It
>> >> >> >> > would
>> >> >> >> > be better if the blocks are kept as they are in the zoom of
>> >> >> >> > 500Kbp
>> >> >> >> > (all
>> >> >> >> > of
>> >> >> >> > the blocks are at one line, shown in the picture2.png). Could
>> >> >> >> > you
>> >> >> >> > please
>> >> >> >> > tell me how to do this? Many thanks!
>> >> >> >> >
>> >> >> >> > BTW: my configure script is like this:
>> >> >> >> >
>> >> >> >> > [methySW_MSH1_1000K]
>> >> >> >> > feature         = methySW_MSH1_1000K
>> >> >> >> > glyph           = xyplot
>> >> >> >> > graph_type      = boxes
>> >> >> >> > fgcolor         = red
>> >> >> >> > bgcolor         = red
>> >> >> >> > height          = 80
>> >> >> >> > min_score       = 0
>> >> >> >> > max_score       = 1
>> >> >> >> > scale           = none
>> >> >> >> > group_on        = methySW_MSH1_1000K
>> >> >> >> > category        = Quantitative Data
>> >> >> >> > label           = 0
>> >> >> >> > key             = Methylation Profile - MSH1_dr_vs_Col0
>> >> >> >> > (Sliding
>> >> >> >> > Windows:1000K-size)
>> >> >> >> >
>> >> >> >> >
>> >> >> >> > Best wishes,
>> >> >> >> > Jiantao
>> >> >> >>
>> >> >> >>
>> >> >> >>
>> >> >> >> --
>> >> >> >>
>> >> >> >>
>> >> >> >>
>> >> >> >> ------------------------------------------------------------------------
>> >> >> >> Scott Cain, Ph. D.                                   scott at
>> >> >> >> scottcain
>> >> >> >> dot net
>> >> >> >> GMOD Coordinator (http://gmod.org/)
>> >> >> >> <a href="tel:216-392-3087" value="+12163923087" target="_blank">216-392-3087
>> >> >> >> Ontario Institute for Cancer Research
>> >> >> >
>> >> >> >
>> >> >>
>> >> >>
>> >> >>
>> >> >> --
>> >> >>
>> >> >>
>> >> >> ------------------------------------------------------------------------
>> >> >> Scott Cain, Ph. D.                                   scott at
>> >> >> scottcain
>> >> >> dot net
>> >> >> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087" target="_blank">216-392-3087
>> >> >> Ontario Institute for Cancer Research
>> >> >
>> >> >
>> >>
>> >>
>> >>
>> >> --
>> >>
>> >> ------------------------------------------------------------------------
>> >> Scott Cain, Ph. D.                                   scott at scottcain
>> >> dot net
>> >> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087" target="_blank">216-392-3087
>> >> Ontario Institute for Cancer Research
>> >
>> >
>>
>>
>>
>> --
>> ------------------------------------------------------------------------
>> Scott Cain, Ph. D.                                   scott at scottcain
>> dot net
>> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087" target="_blank">216-392-3087
>> Ontario Institute for Cancer Research
>
>



--
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087" target="_blank">216-392-3087
Ontario Institute for Cancer Research
Hi



------------------------------------------------------------------------------
One dashboard for servers and applications across Physical-Virtual-Cloud
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
_______________________________________________
Gmod-gbrowse mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse
Reply | Threaded
Open this post in threaded view
|

Re: [Gmod-help] Can I ask a question of Gbrowse2?

Jiantao Yu-2
Hi Scott

I got the error as following, when I tried to install GBrowser via:

Could not find apache executable on this system. Can't figure out version number for config file. at /home/jiantao/GBrowse/install_util/GBrowseInstall.pm line 932.


I tried to solve this problem by
http://sourceforge.net/p/gmod/mailman/message/29595042/

but it didn't help. Can you help me? Thanks!

Jiantao

2015-05-12 0:22 GMT+08:00 Jiantao Yu <[hidden email]>:
Hi Scott

I got the error as following, when I tried to install GBrowser via:

Could not find apache executable on this system. Can't figure out version number for config file. at /home/jiantao/GBrowse/install_util/GBrowseInstall.pm line 932.


I tried to solve this problem by
http://sourceforge.net/p/gmod/mailman/message/29595042/

but it didn't help. Can you help me? Thanks!

Jiantao

2012-04-28 7:53 GMT+08:00 Jiantao Yu <[hidden email]>:
Hi Scott,

Many thanks for your help! It has been successful!! Could you please tell me the reason why we can give the same name to many entries, however, shouldn't give the same IDs to many entries. Except the reason you said that a big bulk of data will belong to one feature in this case, is there any other reason for not doing this? I am not sure why the performance will become poor?

By the way, in the case of my data, I want to group one kind of name, fake, into subtrack, and group another kind of name, fake2, into another subtrack, and give them different colors to display. I used this script, but it looked useless. Could you please help me again? thanks for your patience.

[solveScatter3]
feature         = scatter_feature3
glyph           = xyplot
graph_type      = boxes
subtrack select = Strand strand
subtrack table  = :OverMethy +1 ;
                              :UnderMethy -1
subtrack select fgcolor = sub {
                    my $f = shift;
                    my ($name) = $f->attributes('Name');
                    return "hotpink" if($name =~ /fake/);
                    return "blue" if($name =~ /fake1/);
                }
subtrack select bgcolor = sub {
                    my $f = shift;
                    my ($name) = $f->attributes('Name');
                    return "hotpink" if($name =~ /fake/);
                    return "blue" if($name =~ /fake1/);
                }
label        = 0
height          = 80
min_score       = 0
max_score       = 205
scale           = none
group_on        = display_name
category        = Quantitative Data
key             = solveScatter3

The last question is, why can 'scale' NOT be set up as 'left' or 'right', or else, there are many scale-coordinates (y-axis) appearing in the screen?

Thanks again.

 
Jiantao

2012/4/27 Scott Cain <[hidden email]>
Hi Jiantao,

I managed to reproduce your problem.  It turns out that "group on =
source" doesn't appear to work with the current bioperl.  I fixed it
for the sample data you gave me (for Chr1 only, and one source only),
by adding to each line a "Name=fake;" and then changing the config to
do "group on = display_name".  So, for example, one line of the GFF
would look like this:

Chr1    MSH1_dr_vs_Col_0   meylaSW_MSH1_1000K_feature   475006  524995
 0.777166    .   .   Name=fake;Note=Wi..

Scott


On Fri, Apr 27, 2012 at 12:51 PM, Jiantao Yu <[hidden email]> wrote:
> Hi Scott,
>
> I think I am using Gbrowse2, and I made the configure file based on your
> instruction file, Generic Genome Browser Version 2: A Tutorial for
> Administrators.
>
> However, I am not 100 percent sure of this. Can you tell me how can I check
> the version of Gbrowse.
>
> Another point is that, you see, GBrowse2 didn't provide the configure file
> of Arabidopsis, so we generated this file by ourselves. However, the file is
> made by my colleague, so I am not sure if he took this configure file from
> the GBrowse1.x or Gbrowse2.x . If you are sure the configure file is from
> GBrowse1.x, I will ask my colleague and make sure of this. Thanks.
>
>
> Jiantao
>
> 2012/4/27 Scott Cain <[hidden email]>
>>
>> Hi Jiantao,
>>
>> I forgot to ask--what version of GBrowse are you using?  Your conf
>> file looks like GBrowse 1.X.  Also, I trimmed the GMOD help desk email
>> off the cc list.
>>
>> For xyplots: having the boundaries agacent shouldn't matter; that is
>> what the data in the tutorial are, and it works fine.  There is a
>> problem either with your data, your config or both.  I'm working on a
>> simplified case to see if I can reproduce your problem.
>>
>> For the key/track title problem, I'm guessing you had illigal characters.
>>
>> Scott
>>
>>
>> On Fri, Apr 27, 2012 at 12:13 PM, Jiantao Yu <[hidden email]>
>> wrote:
>> > Hi Scott,
>> >
>> > Thanks for reply. I found some answers for the three questions:
>> > i) The reason that the boxes are displayed in hierarchy, I think, would
>> > be
>> > the boxes are too close. For example, if the start and end positions are
>> > like this:
>> >
>> > 1    10    0.8    .    .
>> > 11    20    0.7    .    .
>> > 21    30    0.7    .    .
>> > ...
>> >
>> > They would probably be in hierarchy when I zoom in. However, if I
>> > changed
>> > the positions like this (+ or - 4):
>> >
>> > 4    7    0.8    .    .
>> > 14    17    0.7    .    .
>> > 24    27    0.7    .    .
>> >
>> > They would be on the same line. However, in the case of my data, it is
>> > hard
>> > to determine what values should be added or subtracted. For example, I
>> > used
>> > 30 as the value, it wasn't enough, I mean, 30 is not enough to ensure
>> > all of
>> > the boxes are on the same line for every scale of zooming in or zooming
>> > out.
>> >
>> > From another point, I really hope the width of boxes are still kept
>> > continuous, you see, no gaps.
>> >
>> > ii) I tried to gave all of the entries, located in the same chromosome,
>> > the
>> > same IDs. This method did solve the problem, i.e. the boxes are fixed on
>> > the
>> > same line. However, as you told in the last email, this led to a
>> > problem,
>> > the performance became poor, and the gbrowse couldn't display the curves
>> > sometimes, if there were lots of boxes which had the same IDs.
>> >
>> > So I think the best way would be, can we find a way to display them on
>> > the
>> > same line, while no limitation of their IDs?
>> >
>> > iii) The reason that there is no title of some tracks, is not very
>> > clear.
>> > However, when I changed the 'key' with some other words, the title can
>> > be
>> > normally displayed. So this has not been a problem now.
>> >
>> > Thanks very much for your patience.
>> >
>> > Regards
>> > Jiantao
>> >
>> >
>> >
>> > 2012/4/27 Scott Cain <[hidden email]>
>> >>
>> >> Jiantao,
>> >>
>> >> Please always respond to the mailing list too (reply-all).
>> >>
>> >> No, you absolutely should not try to group these features by ID.  That
>> >> will result in the database loader putting all of the features
>> >> together into one gigantic object in the database, and your
>> >> performance will be very poor (GBrowse may stop rendering those tracks
>> >> altogether).  I'll try to reproduce your problem now.
>> >>
>> >> Scott
>> >>
>> >>
>> >> On Tue, Apr 24, 2012 at 12:23 PM, Jiantao Yu
>> >> <[hidden email]>
>> >> wrote:
>> >> > Hi Scott,
>> >> >
>> >> > I still got the picture like the attahed file, although I changed
>> >> > group_on=subject. Do you think I need to add 'ID=1' for all of the
>> >> > entries
>> >> > in .gff3 file, in order to let them displayed on the single line?
>> >> >
>> >> > GBrowse may group entries according to their IDs? However, when I did
>> >> > this,
>> >> > I can't get the specific balloon for each position/block in the
>> >> > picture,
>> >> > you
>> >> > see, I just got a very general description in the balloon, such as
>> >> > this
>> >> > position is from 1..30Mbp, which is not what I want to get.
>> >> >
>> >> > Thanks a lot.
>> >> >
>> >> >
>> >> > Jiantao
>> >> >
>> >> > 2012/4/24 Scott Cain <[hidden email]>
>> >> >>
>> >> >> Hi Jiantao,
>> >> >>
>> >> >> "subject" should work for your data.  Did you try panning or zooming
>> >> >> to another location to make sure GBrowse wasn't using a cached image
>> >> >> (just reloading isn't enough)?
>> >> >>
>> >> >> Scott
>> >> >>
>> >> >>
>> >> >> On Tue, Apr 24, 2012 at 11:44 AM, Jiantao Yu
>> >> >> <[hidden email]>
>> >> >> wrote:
>> >> >> > Hi Scott,
>> >> >> >
>> >> >> > Really thanks for you quick reply. I tried to change group-on in
>> >> >> > the
>> >> >> > config
>> >> >> > file into
>> >> >> > group_on = display_name                    or
>> >> >> > group_on =subject
>> >> >> >
>> >> >> > However, it seems no effect. I have attached the .gff3 and .conf
>> >> >> > files,
>> >> >> > and
>> >> >> > the configure script for the .gff3 data starts at the line of
>> >> >> > No.1365.
>> >> >> >
>> >> >> > Could you please have a look at them? Thanks very much.
>> >> >> >
>> >> >> > Best regards
>> >> >> > Jiantao
>> >> >> >
>> >> >> >
>> >> >> > 2012/4/24 Scott Cain <[hidden email]>
>> >> >> >>
>> >> >> >> Hello Jiantao,
>> >> >> >>
>> >> >> >> What you are seeing is almost always a result of the config and
>> >> >> >> the
>> >> >> >> data not matching.  I think your "group_on" setting is incorrect;
>> >> >> >> it
>> >> >> >> is usually something like "group_on = display_name" or "group_on
>> >> >> >> =
>> >> >> >> subject".  I don't think setting it equal to the name of the
>> >> >> >> feature
>> >> >> >> type does anything.  If fixing that doesn't work, you could send
>> >> >> >> a
>> >> >> >> sample of the GFF to the list.  Also, there is a section of the
>> >> >> >> tutorial that came with GBrowse that covers xyplots that might
>> >> >> >> help
>> >> >> >> (look for "quantitative data" in the tutorial).
>> >> >> >>
>> >> >> >> Scott
>> >> >> >>
>> >> >> >>
>> >> >> >> On Mon, Apr 23, 2012 at 8:14 PM, Jiantao Yu
>> >> >> >> <[hidden email]>
>> >> >> >> wrote:
>> >> >> >> > Dear Sir/Madam,
>> >> >> >> >
>> >> >> >> > I highly appreciate the smart of GBrowse. When I tried to
>> >> >> >> > display
>> >> >> >> > the
>> >> >> >> > quantitative data, and set up the zoom with small values, such
>> >> >> >> > as
>> >> >> >> > 40Kbp,
>> >> >> >> > I
>> >> >> >> > got the picture as is attached (picture1.png). However, I
>> >> >> >> > perfer
>> >> >> >> > the
>> >> >> >> > blocks
>> >> >> >> > to locate on the same line, rather than being hierarchical. I
>> >> >> >> > mean,
>> >> >> >> > It
>> >> >> >> > would
>> >> >> >> > be better if the blocks are kept as they are in the zoom of
>> >> >> >> > 500Kbp
>> >> >> >> > (all
>> >> >> >> > of
>> >> >> >> > the blocks are at one line, shown in the picture2.png). Could
>> >> >> >> > you
>> >> >> >> > please
>> >> >> >> > tell me how to do this? Many thanks!
>> >> >> >> >
>> >> >> >> > BTW: my configure script is like this:
>> >> >> >> >
>> >> >> >> > [methySW_MSH1_1000K]
>> >> >> >> > feature         = methySW_MSH1_1000K
>> >> >> >> > glyph           = xyplot
>> >> >> >> > graph_type      = boxes
>> >> >> >> > fgcolor         = red
>> >> >> >> > bgcolor         = red
>> >> >> >> > height          = 80
>> >> >> >> > min_score       = 0
>> >> >> >> > max_score       = 1
>> >> >> >> > scale           = none
>> >> >> >> > group_on        = methySW_MSH1_1000K
>> >> >> >> > category        = Quantitative Data
>> >> >> >> > label           = 0
>> >> >> >> > key             = Methylation Profile - MSH1_dr_vs_Col0
>> >> >> >> > (Sliding
>> >> >> >> > Windows:1000K-size)
>> >> >> >> >
>> >> >> >> >
>> >> >> >> > Best wishes,
>> >> >> >> > Jiantao
>> >> >> >>
>> >> >> >>
>> >> >> >>
>> >> >> >> --
>> >> >> >>
>> >> >> >>
>> >> >> >>
>> >> >> >> ------------------------------------------------------------------------
>> >> >> >> Scott Cain, Ph. D.                                   scott at
>> >> >> >> scottcain
>> >> >> >> dot net
>> >> >> >> GMOD Coordinator (http://gmod.org/)
>> >> >> >> <a href="tel:216-392-3087" value="+12163923087" target="_blank">216-392-3087
>> >> >> >> Ontario Institute for Cancer Research
>> >> >> >
>> >> >> >
>> >> >>
>> >> >>
>> >> >>
>> >> >> --
>> >> >>
>> >> >>
>> >> >> ------------------------------------------------------------------------
>> >> >> Scott Cain, Ph. D.                                   scott at
>> >> >> scottcain
>> >> >> dot net
>> >> >> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087" target="_blank">216-392-3087
>> >> >> Ontario Institute for Cancer Research
>> >> >
>> >> >
>> >>
>> >>
>> >>
>> >> --
>> >>
>> >> ------------------------------------------------------------------------
>> >> Scott Cain, Ph. D.                                   scott at scottcain
>> >> dot net
>> >> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087" target="_blank">216-392-3087
>> >> Ontario Institute for Cancer Research
>> >
>> >
>>
>>
>>
>> --
>> ------------------------------------------------------------------------
>> Scott Cain, Ph. D.                                   scott at scottcain
>> dot net
>> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087" target="_blank">216-392-3087
>> Ontario Institute for Cancer Research
>
>



--
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087" target="_blank">216-392-3087
Ontario Institute for Cancer Research
Hi




------------------------------------------------------------------------------
One dashboard for servers and applications across Physical-Virtual-Cloud
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
_______________________________________________
Gmod-gbrowse mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse
Reply | Threaded
Open this post in threaded view
|

Re: [Gmod-help] Can I ask a question of Gbrowse2?

Scott Cain
Hi Jiantao,

What OS are you using and where is apache installed?

Scott


On Mon, May 11, 2015 at 12:23 PM, Jiantao Yu <[hidden email]> wrote:
Hi Scott

I got the error as following, when I tried to install GBrowser via:

Could not find apache executable on this system. Can't figure out version number for config file. at /home/jiantao/GBrowse/install_util/GBrowseInstall.pm line 932.


I tried to solve this problem by
http://sourceforge.net/p/gmod/mailman/message/29595042/

but it didn't help. Can you help me? Thanks!

Jiantao

2015-05-12 0:22 GMT+08:00 Jiantao Yu <[hidden email]>:
Hi Scott

I got the error as following, when I tried to install GBrowser via:

Could not find apache executable on this system. Can't figure out version number for config file. at /home/jiantao/GBrowse/install_util/GBrowseInstall.pm line 932.


I tried to solve this problem by
http://sourceforge.net/p/gmod/mailman/message/29595042/

but it didn't help. Can you help me? Thanks!

Jiantao

2012-04-28 7:53 GMT+08:00 Jiantao Yu <[hidden email]>:
Hi Scott,

Many thanks for your help! It has been successful!! Could you please tell me the reason why we can give the same name to many entries, however, shouldn't give the same IDs to many entries. Except the reason you said that a big bulk of data will belong to one feature in this case, is there any other reason for not doing this? I am not sure why the performance will become poor?

By the way, in the case of my data, I want to group one kind of name, fake, into subtrack, and group another kind of name, fake2, into another subtrack, and give them different colors to display. I used this script, but it looked useless. Could you please help me again? thanks for your patience.

[solveScatter3]
feature         = scatter_feature3
glyph           = xyplot
graph_type      = boxes
subtrack select = Strand strand
subtrack table  = :OverMethy +1 ;
                              :UnderMethy -1
subtrack select fgcolor = sub {
                    my $f = shift;
                    my ($name) = $f->attributes('Name');
                    return "hotpink" if($name =~ /fake/);
                    return "blue" if($name =~ /fake1/);
                }
subtrack select bgcolor = sub {
                    my $f = shift;
                    my ($name) = $f->attributes('Name');
                    return "hotpink" if($name =~ /fake/);
                    return "blue" if($name =~ /fake1/);
                }
label        = 0
height          = 80
min_score       = 0
max_score       = 205
scale           = none
group_on        = display_name
category        = Quantitative Data
key             = solveScatter3

The last question is, why can 'scale' NOT be set up as 'left' or 'right', or else, there are many scale-coordinates (y-axis) appearing in the screen?

Thanks again.

 
Jiantao

2012/4/27 Scott Cain <[hidden email]>
Hi Jiantao,

I managed to reproduce your problem.  It turns out that "group on =
source" doesn't appear to work with the current bioperl.  I fixed it
for the sample data you gave me (for Chr1 only, and one source only),
by adding to each line a "Name=fake;" and then changing the config to
do "group on = display_name".  So, for example, one line of the GFF
would look like this:

Chr1    MSH1_dr_vs_Col_0   meylaSW_MSH1_1000K_feature   475006  524995
 0.777166    .   .   Name=fake;Note=Wi..

Scott


On Fri, Apr 27, 2012 at 12:51 PM, Jiantao Yu <[hidden email]> wrote:
> Hi Scott,
>
> I think I am using Gbrowse2, and I made the configure file based on your
> instruction file, Generic Genome Browser Version 2: A Tutorial for
> Administrators.
>
> However, I am not 100 percent sure of this. Can you tell me how can I check
> the version of Gbrowse.
>
> Another point is that, you see, GBrowse2 didn't provide the configure file
> of Arabidopsis, so we generated this file by ourselves. However, the file is
> made by my colleague, so I am not sure if he took this configure file from
> the GBrowse1.x or Gbrowse2.x . If you are sure the configure file is from
> GBrowse1.x, I will ask my colleague and make sure of this. Thanks.
>
>
> Jiantao
>
> 2012/4/27 Scott Cain <[hidden email]>
>>
>> Hi Jiantao,
>>
>> I forgot to ask--what version of GBrowse are you using?  Your conf
>> file looks like GBrowse 1.X.  Also, I trimmed the GMOD help desk email
>> off the cc list.
>>
>> For xyplots: having the boundaries agacent shouldn't matter; that is
>> what the data in the tutorial are, and it works fine.  There is a
>> problem either with your data, your config or both.  I'm working on a
>> simplified case to see if I can reproduce your problem.
>>
>> For the key/track title problem, I'm guessing you had illigal characters.
>>
>> Scott
>>
>>
>> On Fri, Apr 27, 2012 at 12:13 PM, Jiantao Yu <[hidden email]>
>> wrote:
>> > Hi Scott,
>> >
>> > Thanks for reply. I found some answers for the three questions:
>> > i) The reason that the boxes are displayed in hierarchy, I think, would
>> > be
>> > the boxes are too close. For example, if the start and end positions are
>> > like this:
>> >
>> > 1    10    0.8    .    .
>> > 11    20    0.7    .    .
>> > 21    30    0.7    .    .
>> > ...
>> >
>> > They would probably be in hierarchy when I zoom in. However, if I
>> > changed
>> > the positions like this (+ or - 4):
>> >
>> > 4    7    0.8    .    .
>> > 14    17    0.7    .    .
>> > 24    27    0.7    .    .
>> >
>> > They would be on the same line. However, in the case of my data, it is
>> > hard
>> > to determine what values should be added or subtracted. For example, I
>> > used
>> > 30 as the value, it wasn't enough, I mean, 30 is not enough to ensure
>> > all of
>> > the boxes are on the same line for every scale of zooming in or zooming
>> > out.
>> >
>> > From another point, I really hope the width of boxes are still kept
>> > continuous, you see, no gaps.
>> >
>> > ii) I tried to gave all of the entries, located in the same chromosome,
>> > the
>> > same IDs. This method did solve the problem, i.e. the boxes are fixed on
>> > the
>> > same line. However, as you told in the last email, this led to a
>> > problem,
>> > the performance became poor, and the gbrowse couldn't display the curves
>> > sometimes, if there were lots of boxes which had the same IDs.
>> >
>> > So I think the best way would be, can we find a way to display them on
>> > the
>> > same line, while no limitation of their IDs?
>> >
>> > iii) The reason that there is no title of some tracks, is not very
>> > clear.
>> > However, when I changed the 'key' with some other words, the title can
>> > be
>> > normally displayed. So this has not been a problem now.
>> >
>> > Thanks very much for your patience.
>> >
>> > Regards
>> > Jiantao
>> >
>> >
>> >
>> > 2012/4/27 Scott Cain <[hidden email]>
>> >>
>> >> Jiantao,
>> >>
>> >> Please always respond to the mailing list too (reply-all).
>> >>
>> >> No, you absolutely should not try to group these features by ID.  That
>> >> will result in the database loader putting all of the features
>> >> together into one gigantic object in the database, and your
>> >> performance will be very poor (GBrowse may stop rendering those tracks
>> >> altogether).  I'll try to reproduce your problem now.
>> >>
>> >> Scott
>> >>
>> >>
>> >> On Tue, Apr 24, 2012 at 12:23 PM, Jiantao Yu
>> >> <[hidden email]>
>> >> wrote:
>> >> > Hi Scott,
>> >> >
>> >> > I still got the picture like the attahed file, although I changed
>> >> > group_on=subject. Do you think I need to add 'ID=1' for all of the
>> >> > entries
>> >> > in .gff3 file, in order to let them displayed on the single line?
>> >> >
>> >> > GBrowse may group entries according to their IDs? However, when I did
>> >> > this,
>> >> > I can't get the specific balloon for each position/block in the
>> >> > picture,
>> >> > you
>> >> > see, I just got a very general description in the balloon, such as
>> >> > this
>> >> > position is from 1..30Mbp, which is not what I want to get.
>> >> >
>> >> > Thanks a lot.
>> >> >
>> >> >
>> >> > Jiantao
>> >> >
>> >> > 2012/4/24 Scott Cain <[hidden email]>
>> >> >>
>> >> >> Hi Jiantao,
>> >> >>
>> >> >> "subject" should work for your data.  Did you try panning or zooming
>> >> >> to another location to make sure GBrowse wasn't using a cached image
>> >> >> (just reloading isn't enough)?
>> >> >>
>> >> >> Scott
>> >> >>
>> >> >>
>> >> >> On Tue, Apr 24, 2012 at 11:44 AM, Jiantao Yu
>> >> >> <[hidden email]>
>> >> >> wrote:
>> >> >> > Hi Scott,
>> >> >> >
>> >> >> > Really thanks for you quick reply. I tried to change group-on in
>> >> >> > the
>> >> >> > config
>> >> >> > file into
>> >> >> > group_on = display_name                    or
>> >> >> > group_on =subject
>> >> >> >
>> >> >> > However, it seems no effect. I have attached the .gff3 and .conf
>> >> >> > files,
>> >> >> > and
>> >> >> > the configure script for the .gff3 data starts at the line of
>> >> >> > No.1365.
>> >> >> >
>> >> >> > Could you please have a look at them? Thanks very much.
>> >> >> >
>> >> >> > Best regards
>> >> >> > Jiantao
>> >> >> >
>> >> >> >
>> >> >> > 2012/4/24 Scott Cain <[hidden email]>
>> >> >> >>
>> >> >> >> Hello Jiantao,
>> >> >> >>
>> >> >> >> What you are seeing is almost always a result of the config and
>> >> >> >> the
>> >> >> >> data not matching.  I think your "group_on" setting is incorrect;
>> >> >> >> it
>> >> >> >> is usually something like "group_on = display_name" or "group_on
>> >> >> >> =
>> >> >> >> subject".  I don't think setting it equal to the name of the
>> >> >> >> feature
>> >> >> >> type does anything.  If fixing that doesn't work, you could send
>> >> >> >> a
>> >> >> >> sample of the GFF to the list.  Also, there is a section of the
>> >> >> >> tutorial that came with GBrowse that covers xyplots that might
>> >> >> >> help
>> >> >> >> (look for "quantitative data" in the tutorial).
>> >> >> >>
>> >> >> >> Scott
>> >> >> >>
>> >> >> >>
>> >> >> >> On Mon, Apr 23, 2012 at 8:14 PM, Jiantao Yu
>> >> >> >> <[hidden email]>
>> >> >> >> wrote:
>> >> >> >> > Dear Sir/Madam,
>> >> >> >> >
>> >> >> >> > I highly appreciate the smart of GBrowse. When I tried to
>> >> >> >> > display
>> >> >> >> > the
>> >> >> >> > quantitative data, and set up the zoom with small values, such
>> >> >> >> > as
>> >> >> >> > 40Kbp,
>> >> >> >> > I
>> >> >> >> > got the picture as is attached (picture1.png). However, I
>> >> >> >> > perfer
>> >> >> >> > the
>> >> >> >> > blocks
>> >> >> >> > to locate on the same line, rather than being hierarchical. I
>> >> >> >> > mean,
>> >> >> >> > It
>> >> >> >> > would
>> >> >> >> > be better if the blocks are kept as they are in the zoom of
>> >> >> >> > 500Kbp
>> >> >> >> > (all
>> >> >> >> > of
>> >> >> >> > the blocks are at one line, shown in the picture2.png). Could
>> >> >> >> > you
>> >> >> >> > please
>> >> >> >> > tell me how to do this? Many thanks!
>> >> >> >> >
>> >> >> >> > BTW: my configure script is like this:
>> >> >> >> >
>> >> >> >> > [methySW_MSH1_1000K]
>> >> >> >> > feature         = methySW_MSH1_1000K
>> >> >> >> > glyph           = xyplot
>> >> >> >> > graph_type      = boxes
>> >> >> >> > fgcolor         = red
>> >> >> >> > bgcolor         = red
>> >> >> >> > height          = 80
>> >> >> >> > min_score       = 0
>> >> >> >> > max_score       = 1
>> >> >> >> > scale           = none
>> >> >> >> > group_on        = methySW_MSH1_1000K
>> >> >> >> > category        = Quantitative Data
>> >> >> >> > label           = 0
>> >> >> >> > key             = Methylation Profile - MSH1_dr_vs_Col0
>> >> >> >> > (Sliding
>> >> >> >> > Windows:1000K-size)
>> >> >> >> >
>> >> >> >> >
>> >> >> >> > Best wishes,
>> >> >> >> > Jiantao
>> >> >> >>
>> >> >> >>
>> >> >> >>
>> >> >> >> --
>> >> >> >>
>> >> >> >>
>> >> >> >>
>> >> >> >> ------------------------------------------------------------------------
>> >> >> >> Scott Cain, Ph. D.                                   scott at
>> >> >> >> scottcain
>> >> >> >> dot net
>> >> >> >> GMOD Coordinator (http://gmod.org/)
>> >> >> >> <a href="tel:216-392-3087" value="+12163923087" target="_blank">216-392-3087
>> >> >> >> Ontario Institute for Cancer Research
>> >> >> >
>> >> >> >
>> >> >>
>> >> >>
>> >> >>
>> >> >> --
>> >> >>
>> >> >>
>> >> >> ------------------------------------------------------------------------
>> >> >> Scott Cain, Ph. D.                                   scott at
>> >> >> scottcain
>> >> >> dot net
>> >> >> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087" target="_blank">216-392-3087
>> >> >> Ontario Institute for Cancer Research
>> >> >
>> >> >
>> >>
>> >>
>> >>
>> >> --
>> >>
>> >> ------------------------------------------------------------------------
>> >> Scott Cain, Ph. D.                                   scott at scottcain
>> >> dot net
>> >> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087" target="_blank">216-392-3087
>> >> Ontario Institute for Cancer Research
>> >
>> >
>>
>>
>>
>> --
>> ------------------------------------------------------------------------
>> Scott Cain, Ph. D.                                   scott at scottcain
>> dot net
>> GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087" target="_blank">216-392-3087
>> Ontario Institute for Cancer Research
>
>



--
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     <a href="tel:216-392-3087" value="+12163923087" target="_blank">216-392-3087
Ontario Institute for Cancer Research
Hi






--
------------------------------------------------------------------------
Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (http://gmod.org/)                     216-392-3087
Ontario Institute for Cancer Research

------------------------------------------------------------------------------
One dashboard for servers and applications across Physical-Virtual-Cloud
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
_______________________________________________
Gmod-gbrowse mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse