[Gmod-ajax] IndexedFasta?

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

[Gmod-ajax] IndexedFasta?

Ben Bimber
Hello,

Sorry if I missed documentation somewhere on this; however, when looking over the code perl / JS code for preparing sequences, I came across the IndexedFasta / --indexed_fasta options.  If not clear on what this does - can we now load FASTA tracks directly, w/o needing to process them to JSON?  

-Ben

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gmod-ajax mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-ajax
Reply | Threaded
Open this post in threaded view
|

Re: IndexedFasta?

Colin
Yep, this has been available since 1.12.0. The concept of an "indexed fasta" refers to running "samtools faidx yourfile.fa" to create yourfile.fa.fai. Then you can use prepare-refseqs on it a la prepare-refseqs.pl --indexed_fasta yourfile.fa

This will copy your indexed fasta into the data directory with a track entry in trackList.json (note that you might want to add "useAsRefSeqStore": true to the refseq track in trackList.json in 1.12.1 or 1.12.0, but that is fixed on master)

TwoBit file support is also pending!

-Colin

On Wed, Mar 22, 2017 at 9:17 AM, Ben Bimber <[hidden email]> wrote:
Hello,

Sorry if I missed documentation somewhere on this; however, when looking over the code perl / JS code for preparing sequences, I came across the IndexedFasta / --indexed_fasta options.  If not clear on what this does - can we now load FASTA tracks directly, w/o needing to process them to JSON?  

-Ben

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gmod-ajax mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-ajax



------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gmod-ajax mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-ajax
Reply | Threaded
Open this post in threaded view
|

Re: IndexedFasta?

Ben Bimber
Interesting.  It would be nice not to have to duplicate data any more than necessary.  Are there any downsides to this route, as opposed to the old route of JSONifying the sequence?

For other tabix formats like BED, GFF, etc, do you expect similar options to exist at some point?  

On Wed, Mar 22, 2017 at 8:05 AM, Colin <[hidden email]> wrote:
Yep, this has been available since 1.12.0. The concept of an "indexed fasta" refers to running "samtools faidx yourfile.fa" to create yourfile.fa.fai. Then you can use prepare-refseqs on it a la prepare-refseqs.pl --indexed_fasta yourfile.fa

This will copy your indexed fasta into the data directory with a track entry in trackList.json (note that you might want to add "useAsRefSeqStore": true to the refseq track in trackList.json in 1.12.1 or 1.12.0, but that is fixed on master)

TwoBit file support is also pending!

-Colin

On Wed, Mar 22, 2017 at 9:17 AM, Ben Bimber <[hidden email]> wrote:
Hello,

Sorry if I missed documentation somewhere on this; however, when looking over the code perl / JS code for preparing sequences, I came across the IndexedFasta / --indexed_fasta options.  If not clear on what this does - can we now load FASTA tracks directly, w/o needing to process them to JSON?  

-Ben

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gmod-ajax mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-ajax




------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gmod-ajax mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-ajax
Reply | Threaded
Open this post in threaded view
|

Re: IndexedFasta?

Colin
I don't think there are any downsides to alternative sequence adapters like IndexedFasta/TwoBit, but there could be something I haven't though of yet

For the BED and GFF tabix functionality, do you mean adding a perl loading script for them? And I guess as far as options, there are also a couple places where GFF tabix hasn't caught up full par to flatfile-to-json.pl but that might improve see https://github.com/gmod/jbrowse/issues/780

-Colin

On Wed, Mar 22, 2017 at 12:47 PM, Ben Bimber <[hidden email]> wrote:
Interesting.  It would be nice not to have to duplicate data any more than necessary.  Are there any downsides to this route, as opposed to the old route of JSONifying the sequence?

For other tabix formats like BED, GFF, etc, do you expect similar options to exist at some point?  

On Wed, Mar 22, 2017 at 8:05 AM, Colin <[hidden email]> wrote:
Yep, this has been available since 1.12.0. The concept of an "indexed fasta" refers to running "samtools faidx yourfile.fa" to create yourfile.fa.fai. Then you can use prepare-refseqs on it a la prepare-refseqs.pl --indexed_fasta yourfile.fa

This will copy your indexed fasta into the data directory with a track entry in trackList.json (note that you might want to add "useAsRefSeqStore": true to the refseq track in trackList.json in 1.12.1 or 1.12.0, but that is fixed on master)

TwoBit file support is also pending!

-Colin

On Wed, Mar 22, 2017 at 9:17 AM, Ben Bimber <[hidden email]> wrote:
Hello,

Sorry if I missed documentation somewhere on this; however, when looking over the code perl / JS code for preparing sequences, I came across the IndexedFasta / --indexed_fasta options.  If not clear on what this does - can we now load FASTA tracks directly, w/o needing to process them to JSON?  

-Ben

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gmod-ajax mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-ajax





------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gmod-ajax mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-ajax
Reply | Threaded
Open this post in threaded view
|

Re: IndexedFasta?

Ben Bimber
The last time I looked into capabilities, which I think was within the year, but apparently this is out of date, directly reading GFFs required loading the entire file into the client, as opposed to downloading/parsing windows.  Your link suggests this is changed though.  

I'm especially interested in directly reading the raw data because 1) it means no parse step and duplication, and 2) so long as one can generate the track JSON, it removes all the perl dependencies.

i'll check that out.

-Ben

On Wed, Mar 22, 2017 at 4:10 PM, Colin <[hidden email]> wrote:
I don't think there are any downsides to alternative sequence adapters like IndexedFasta/TwoBit, but there could be something I haven't though of yet

For the BED and GFF tabix functionality, do you mean adding a perl loading script for them? And I guess as far as options, there are also a couple places where GFF tabix hasn't caught up full par to flatfile-to-json.pl but that might improve see https://github.com/gmod/jbrowse/issues/780

-Colin

On Wed, Mar 22, 2017 at 12:47 PM, Ben Bimber <[hidden email]> wrote:
Interesting.  It would be nice not to have to duplicate data any more than necessary.  Are there any downsides to this route, as opposed to the old route of JSONifying the sequence?

For other tabix formats like BED, GFF, etc, do you expect similar options to exist at some point?  

On Wed, Mar 22, 2017 at 8:05 AM, Colin <[hidden email]> wrote:
Yep, this has been available since 1.12.0. The concept of an "indexed fasta" refers to running "samtools faidx yourfile.fa" to create yourfile.fa.fai. Then you can use prepare-refseqs on it a la prepare-refseqs.pl --indexed_fasta yourfile.fa

This will copy your indexed fasta into the data directory with a track entry in trackList.json (note that you might want to add "useAsRefSeqStore": true to the refseq track in trackList.json in 1.12.1 or 1.12.0, but that is fixed on master)

TwoBit file support is also pending!

-Colin

On Wed, Mar 22, 2017 at 9:17 AM, Ben Bimber <[hidden email]> wrote:
Hello,

Sorry if I missed documentation somewhere on this; however, when looking over the code perl / JS code for preparing sequences, I came across the IndexedFasta / --indexed_fasta options.  If not clear on what this does - can we now load FASTA tracks directly, w/o needing to process them to JSON?  

-Ben

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gmod-ajax mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-ajax






------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gmod-ajax mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gmod-ajax