blastp alignments not shown (analysis blast module)

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

blastp alignments not shown (analysis blast module)

Matthew Hobbs-2
I am using the tripal blast analysis module (6.x-1.1-tripal_v1.1) and have been loading in some blastp results but found alignments were not being displayed. This is because the module tests for existence of alignments by using the 'query_frame' attribute of the hsp (which is derived from the XML 'Hsp_query-frame' element):

lines 1037-1038 tripal_blast_analysis: grep -n -B1 query_frame includes/parse_blast_XML.inc :

                  // if there is at least one HSP
                  if (!empty($hsp_array[0]['query_frame'])) {

lines 88-89 theme/tripal_feature/tripal_feature_blast_results.tpl.php :

           <!-- If there is alignment information for at least one HSP -->
           <?php if (!empty($hit['hsp'][0]['query_frame'])) { ?>

The query in blastp is a protein sequence for which frame is not applicable and so there is no 'Hsp_query-frame' element in the XML (and I guess the same is true for tblastn but haven't checked). When I change the above lines to use 'midline' instead of 'query_frame' then my alignments are available.


Matthew

Book Online tyrannosaurs.com.au

Click here to read the Australian Museum email disclaimer:

 http://info.austmus.gov.au/disclaimer.htm



------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
_______________________________________________
Gmod-tripal mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-tripal
Reply | Threaded
Open this post in threaded view
|

Re: blastp alignments not shown (analysis blast module)

Michael Dondrup-3
Hi Mathew,

I have recently imported blastp vs NR results into Tripal, and they seem to have been imported correctly. However, I haven't checked
for the presence of all results systematically, but all genes I have checked manually which should have results have them. I am now a little concerned that
the import was incomplete and wish to very the import. Therefore, I would like to ask a few questions for clarification (note I am not a tripal developer) if I may.

- What do you mean by:   "but found alignments were not being displayed"? I remember that I was looking for blast hits in the wrong place in the beginning.
The blast hits show up on the feature page links as "<Database> homologs". I didn't see a way to display the blast homology string, only e-value and score
are displayed per hit.

- Do you find all results missing or only some?

- Do you have a way to check the presence of all alignments programmatically independent of the importer?

Best
Michael


Michael Dondrup
Postdoctoral fellow
Sea Lice Research Centre/Department of Informatics
University of Bergen
Thormøhlensgate 55, N-5008 Bergen,
Norway

On Oct 9, 2013, at 11:39 PM, Matthew Hobbs wrote:

> I am using the tripal blast analysis module (6.x-1.1-tripal_v1.1) and have been loading in some blastp results but found alignments were not being displayed. This is because the module tests for existence of alignments by using the 'query_frame' attribute of the hsp (which is derived from the XML 'Hsp_query-frame' element):
>
> lines 1037-1038 tripal_blast_analysis: grep -n -B1 query_frame includes/parse_blast_XML.inc :
>
>                   // if there is at least one HSP
>                   if (!empty($hsp_array[0]['query_frame'])) {
>
> lines 88-89 theme/tripal_feature/tripal_feature_blast_results.tpl.php :
>
>            <!-- If there is alignment information for at least one HSP -->
>            <?php if (!empty($hit['hsp'][0]['query_frame'])) { ?>
>
> The query in blastp is a protein sequence for which frame is not applicable and so there is no 'Hsp_query-frame' element in the XML (and I guess the same is true for tblastn but haven't checked). When I change the above lines to use 'midline' instead of 'query_frame' then my alignments are available.
>
>
> Matthew
>
> Book Online tyrannosaurs.com.au
> Click here to read the Australian Museum email disclaimer:
>  http://info.austmus.gov.au/disclaimer.htm
>
> ------------------------------------------------------------------------------
> October Webinars: Code for Performance
> Free Intel webinars can help you accelerate application performance.
> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from
> the latest Intel processors and coprocessors. See abstracts and register >
> http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk_______________________________________________
> Gmod-tripal mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gmod-tripal

------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
_______________________________________________
Gmod-tripal mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-tripal

signature.asc (465 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: blastp alignments not shown (analysis blast module)

Matthew Hobbs-2
Hi Michael,

Sorry if my post was unclear.

"I didn't see a way to display the blast homology string, only e-value and score" - yes, exactly the problem. Perhaps like you I didn't know exactly what to expect when viewing blast results and was happy to see just a table of hits. However you should also see a sequence alignment see e.g.

http://www.vaccinium.org/ID2848

Note on this page that at the start of each row there are little triangualr icons you can click on to toggle a display of an alignment like this:

HSP 1

Score: 123.25 bits (308), Expect = 8.28308e-27
Identity = 59/61 (96.72%), Postives = 60/61 (98.36%), Query Frame = 1

Query:   64 MGSSEERVVAVIMVGGPTKGTRFRPLSLNIPKPLFPLAGQPMVHHPISACKRIPXLAQIYL 246
            MGSSEERVVAVIMVGGPTKGTRFRPLSLNIPKPLFPLAGQPMVHHPISACK+IP LAQIYL
Sbjct:    1 MGSSEERVVAVIMVGGPTKGTRFRPLSLNIPKPLFPLAGQPMVHHPISACKKIPNLAQIYL 61


Cool - but this feature was missing with my blastp results, hence my investigation.

To answer your questions about missing data -  it's not a problem to do with loading the data, just to do with how the module checks whether any hsp sequence alignment are available for a hit, so data is not missing, just not fully displayed..

Matthew




________________________________________
From: Michael Dondrup [[hidden email]]
Sent: Thursday, 10 October 2013 7:27 PM
To: Matthew Hobbs
Cc: GMOD Tripal ‎[[hidden email]]‎
Subject: Re: [Gmod-tripal] blastp alignments not shown (analysis blast module)

Hi Mathew,

I have recently imported blastp vs NR results into Tripal, and they seem to have been imported correctly. However, I haven't checked
for the presence of all results systematically, but all genes I have checked manually which should have results have them. I am now a little concerned that
the import was incomplete and wish to very the import. Therefore, I would like to ask a few questions for clarification (note I am not a tripal developer) if I may.

- What do you mean by:   "but found alignments were not being displayed"? I remember that I was looking for blast hits in the wrong place in the beginning.
The blast hits show up on the feature page links as "<Database> homologs". I didn't see a way to display the blast homology string, only e-value and score
are displayed per hit.

- Do you find all results missing or only some?

- Do you have a way to check the presence of all alignments programmatically independent of the importer?

Best
Michael


Michael Dondrup
Postdoctoral fellow
Sea Lice Research Centre/Department of Informatics
University of Bergen
Thormøhlensgate 55, N-5008 Bergen,
Norway
TYRANNOSAURS - Meet the family
Exhibition opens 23 November 2013

Book online http://tyrannosaurs.com.au


Click here to read the Australian Museum email disclaimer:
http://info.austmus.gov.au/disclaimer.htm
------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
_______________________________________________
Gmod-tripal mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-tripal
Reply | Threaded
Open this post in threaded view
|

Re: blastp alignments not shown (analysis blast module)

Michael Dondrup-3
Hi Matthew,

thanks for the clarification. This is indeed a cool feature which I also want to activate.

Best
Michael


sorry for misspelling your name, btw.


Michael Dondrup
Postdoctoral fellow
Sea Lice Research Centre/Department of Informatics
University of Bergen
Thormøhlensgate 55, N-5008 Bergen,
Norway

On Oct 10, 2013, at 11:49 AM, Matthew Hobbs wrote:

> Hi Michael,
>
> Sorry if my post was unclear.
>
> "I didn't see a way to display the blast homology string, only e-value and score" - yes, exactly the problem. Perhaps like you I didn't know exactly what to expect when viewing blast results and was happy to see just a table of hits. However you should also see a sequence alignment see e.g.
>
> http://www.vaccinium.org/ID2848
>
> Note on this page that at the start of each row there are little triangualr icons you can click on to toggle a display of an alignment like this:
>
> HSP 1
>
> Score: 123.25 bits (308), Expect = 8.28308e-27
> Identity = 59/61 (96.72%), Postives = 60/61 (98.36%), Query Frame = 1
>
> Query:   64 MGSSEERVVAVIMVGGPTKGTRFRPLSLNIPKPLFPLAGQPMVHHPISACKRIPXLAQIYL 246
>            MGSSEERVVAVIMVGGPTKGTRFRPLSLNIPKPLFPLAGQPMVHHPISACK+IP LAQIYL
> Sbjct:    1 MGSSEERVVAVIMVGGPTKGTRFRPLSLNIPKPLFPLAGQPMVHHPISACKKIPNLAQIYL 61
>
>
> Cool - but this feature was missing with my blastp results, hence my investigation.
>
> To answer your questions about missing data -  it's not a problem to do with loading the data, just to do with how the module checks whether any hsp sequence alignment are available for a hit, so data is not missing, just not fully displayed..
>
> Matthew
>
>
>
>
> ________________________________________
> From: Michael Dondrup [[hidden email]]
> Sent: Thursday, 10 October 2013 7:27 PM
> To: Matthew Hobbs
> Cc: GMOD Tripal ‎[[hidden email]]‎
> Subject: Re: [Gmod-tripal] blastp alignments not shown (analysis blast module)
>
> Hi Mathew,
>
> I have recently imported blastp vs NR results into Tripal, and they seem to have been imported correctly. However, I haven't checked
> for the presence of all results systematically, but all genes I have checked manually which should have results have them. I am now a little concerned that
> the import was incomplete and wish to very the import. Therefore, I would like to ask a few questions for clarification (note I am not a tripal developer) if I may.
>
> - What do you mean by:   "but found alignments were not being displayed"? I remember that I was looking for blast hits in the wrong place in the beginning.
> The blast hits show up on the feature page links as "<Database> homologs". I didn't see a way to display the blast homology string, only e-value and score
> are displayed per hit.
>
> - Do you find all results missing or only some?
>
> - Do you have a way to check the presence of all alignments programmatically independent of the importer?
>
> Best
> Michael
>
>
> Michael Dondrup
> Postdoctoral fellow
> Sea Lice Research Centre/Department of Informatics
> University of Bergen
> Thormøhlensgate 55, N-5008 Bergen,
> Norway
> TYRANNOSAURS - Meet the family
> Exhibition opens 23 November 2013
>
> Book online http://tyrannosaurs.com.au
>
>
> Click here to read the Australian Museum email disclaimer:
> http://info.austmus.gov.au/disclaimer.htm

------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
_______________________________________________
Gmod-tripal mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-tripal

signature.asc (465 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: blastp alignments not shown (analysis blast module)

Michael Dondrup-3
I have just made the suggested changes and I can confirm that I now can see the blast alignments as described.
I think that will be very helpful for our users, and hope that the patch will make it into the next version.

Thanks a ton
Michael

On Oct 10, 2013, at 12:55 PM, Michael Dondrup wrote:

> Hi Matthew,
>
> thanks for the clarification. This is indeed a cool feature which I also want to activate.
>
> Best
> Michael
>
>
> sorry for misspelling your name, btw.
>
>
> Michael Dondrup
> Postdoctoral fellow
> Sea Lice Research Centre/Department of Informatics
> University of Bergen
> Thormøhlensgate 55, N-5008 Bergen,
> Norway
>
> On Oct 10, 2013, at 11:49 AM, Matthew Hobbs wrote:
>
>> Hi Michael,
>>
>> Sorry if my post was unclear.
>>
>> "I didn't see a way to display the blast homology string, only e-value and score" - yes, exactly the problem. Perhaps like you I didn't know exactly what to expect when viewing blast results and was happy to see just a table of hits. However you should also see a sequence alignment see e.g.
>>
>> http://www.vaccinium.org/ID2848
>>
>> Note on this page that at the start of each row there are little triangualr icons you can click on to toggle a display of an alignment like this:
>>
>> HSP 1
>>
>> Score: 123.25 bits (308), Expect = 8.28308e-27
>> Identity = 59/61 (96.72%), Postives = 60/61 (98.36%), Query Frame = 1
>>
>> Query:   64 MGSSEERVVAVIMVGGPTKGTRFRPLSLNIPKPLFPLAGQPMVHHPISACKRIPXLAQIYL 246
>>           MGSSEERVVAVIMVGGPTKGTRFRPLSLNIPKPLFPLAGQPMVHHPISACK+IP LAQIYL
>> Sbjct:    1 MGSSEERVVAVIMVGGPTKGTRFRPLSLNIPKPLFPLAGQPMVHHPISACKKIPNLAQIYL 61
>>
>>
>> Cool - but this feature was missing with my blastp results, hence my investigation.
>>
>> To answer your questions about missing data -  it's not a problem to do with loading the data, just to do with how the module checks whether any hsp sequence alignment are available for a hit, so data is not missing, just not fully displayed..
>>
>> Matthew
>>
>>
>>
>>
>> ________________________________________
>> From: Michael Dondrup [[hidden email]]
>> Sent: Thursday, 10 October 2013 7:27 PM
>> To: Matthew Hobbs
>> Cc: GMOD Tripal ‎[[hidden email]]‎
>> Subject: Re: [Gmod-tripal] blastp alignments not shown (analysis blast module)
>>
>> Hi Mathew,
>>
>> I have recently imported blastp vs NR results into Tripal, and they seem to have been imported correctly. However, I haven't checked
>> for the presence of all results systematically, but all genes I have checked manually which should have results have them. I am now a little concerned that
>> the import was incomplete and wish to very the import. Therefore, I would like to ask a few questions for clarification (note I am not a tripal developer) if I may.
>>
>> - What do you mean by:   "but found alignments were not being displayed"? I remember that I was looking for blast hits in the wrong place in the beginning.
>> The blast hits show up on the feature page links as "<Database> homologs". I didn't see a way to display the blast homology string, only e-value and score
>> are displayed per hit.
>>
>> - Do you find all results missing or only some?
>>
>> - Do you have a way to check the presence of all alignments programmatically independent of the importer?
>>
>> Best
>> Michael
>>
>>
>> Michael Dondrup
>> Postdoctoral fellow
>> Sea Lice Research Centre/Department of Informatics
>> University of Bergen
>> Thormøhlensgate 55, N-5008 Bergen,
>> Norway
>> TYRANNOSAURS - Meet the family
>> Exhibition opens 23 November 2013
>>
>> Book online http://tyrannosaurs.com.au
>>
>>
>> Click here to read the Australian Museum email disclaimer:
>> http://info.austmus.gov.au/disclaimer.htm
>
> ------------------------------------------------------------------------------
> October Webinars: Code for Performance
> Free Intel webinars can help you accelerate application performance.
> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from
> the latest Intel processors and coprocessors. See abstracts and register >
> http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk_______________________________________________
> Gmod-tripal mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gmod-tripal

------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
_______________________________________________
Gmod-tripal mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-tripal

signature.asc (465 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: blastp alignments not shown (analysis blast module)

Stephen Ficklin-2
Thank you Mathew for pointing out the problem, and thank you Michael for confirming the solution.

I have added an issue to the project bug tracking here:  https://drupal.org/node/2109049

If you have a drupal account feel free to subscribe to that page and you'll be notified when this bug is corrected.

Thank much!
Stephen

On 10/10/2013 7:38 AM, Michael Dondrup wrote:
I have just made the suggested changes and I can confirm that I now can see the blast alignments as described.
I think that will be very helpful for our users, and hope that the patch will make it into the next version.

Thanks a ton
Michael 

On Oct 10, 2013, at 12:55 PM, Michael Dondrup wrote:

Hi Matthew,

thanks for the clarification. This is indeed a cool feature which I also want to activate.

Best
Michael


sorry for misspelling your name, btw.


Michael Dondrup
Postdoctoral fellow
Sea Lice Research Centre/Department of Informatics
University of Bergen
Thormøhlensgate 55, N-5008 Bergen, 
Norway

On Oct 10, 2013, at 11:49 AM, Matthew Hobbs wrote:

Hi Michael,

Sorry if my post was unclear.

"I didn't see a way to display the blast homology string, only e-value and score" - yes, exactly the problem. Perhaps like you I didn't know exactly what to expect when viewing blast results and was happy to see just a table of hits. However you should also see a sequence alignment see e.g.

http://www.vaccinium.org/ID2848

Note on this page that at the start of each row there are little triangualr icons you can click on to toggle a display of an alignment like this:

HSP 1

Score: 123.25 bits (308), Expect = 8.28308e-27
Identity = 59/61 (96.72%), Postives = 60/61 (98.36%), Query Frame = 1

Query:   64 MGSSEERVVAVIMVGGPTKGTRFRPLSLNIPKPLFPLAGQPMVHHPISACKRIPXLAQIYL 246
          MGSSEERVVAVIMVGGPTKGTRFRPLSLNIPKPLFPLAGQPMVHHPISACK+IP LAQIYL
Sbjct:    1 MGSSEERVVAVIMVGGPTKGTRFRPLSLNIPKPLFPLAGQPMVHHPISACKKIPNLAQIYL 61


Cool - but this feature was missing with my blastp results, hence my investigation.

To answer your questions about missing data -  it's not a problem to do with loading the data, just to do with how the module checks whether any hsp sequence alignment are available for a hit, so data is not missing, just not fully displayed..

Matthew




________________________________________
From: Michael Dondrup [[hidden email]]
Sent: Thursday, 10 October 2013 7:27 PM
To: Matthew Hobbs
Cc: GMOD Tripal ‎[[hidden email]]‎
Subject: Re: [Gmod-tripal] blastp alignments not shown (analysis blast module)

Hi Mathew,

I have recently imported blastp vs NR results into Tripal, and they seem to have been imported correctly. However, I haven't checked
for the presence of all results systematically, but all genes I have checked manually which should have results have them. I am now a little concerned that
the import was incomplete and wish to very the import. Therefore, I would like to ask a few questions for clarification (note I am not a tripal developer) if I may.

- What do you mean by:   "but found alignments were not being displayed"? I remember that I was looking for blast hits in the wrong place in the beginning.
The blast hits show up on the feature page links as "<Database> homologs". I didn't see a way to display the blast homology string, only e-value and score
are displayed per hit.

- Do you find all results missing or only some?

- Do you have a way to check the presence of all alignments programmatically independent of the importer?

Best
Michael


Michael Dondrup
Postdoctoral fellow
Sea Lice Research Centre/Department of Informatics
University of Bergen
Thormøhlensgate 55, N-5008 Bergen,
Norway
TYRANNOSAURS - Meet the family
Exhibition opens 23 November 2013

Book online http://tyrannosaurs.com.au


Click here to read the Australian Museum email disclaimer:
http://info.austmus.gov.au/disclaimer.htm
------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk_______________________________________________
Gmod-tripal mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-tripal

      

------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk


_______________________________________________
Gmod-tripal mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-tripal


------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
_______________________________________________
Gmod-tripal mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-tripal