WebApollo to chado - CDS?

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

WebApollo to chado - CDS?

Marc Höppner-3
Hi,

been playing around with WebApollo and exports to a Chado db (and then onwards to a Jbrowse portal via biodb-to-json.pl ). I noticed that WebApollo does not seem to export CDS features chad, but a polypeptides instead (?). This is a bit of an issue since without CDS features the transcript is somewhat incomplete - and the CSS formatting in Jbrowse also depends on CDS features to be present in a coding transcript.

Am I missing something perhaps?

/Marc




Reply | Threaded
Open this post in threaded view
|

Re: WebApollo to chado - CDS?

Colin
Hi Marc,

Great idea. I tried testing this out and I setup a round-trip Chado export and biodb-to-json script and I can see what you mean. I don't think the polypeptide is in place of the CDS, but I found another option to get the CDS.

I found that if I added "inferCDS" to the biodb loader on jbrowse, then you can infer the CDS feaures.


diff --git a/src/perl5/Bio/JBrowse/Cmd/BioDBToJson.pm b/src/perl5/Bio/JBrowse/Cmd/BioDBToJson.pm
index d56b273..8dcfc96 100644
--- a/src/perl5/Bio/JBrowse/Cmd/BioDBToJson.pm
+++ b/src/perl5/Bio/JBrowse/Cmd/BioDBToJson.pm
@@ -74,6 +74,7 @@ sub run {
     }
     $db->strict_bounds_checking(1) if $db->can('strict_bounds_checking');
     $db->absolute(1)               if $db->can('absolute');
+    $db->inferCDS(1);
 
     foreach my $seg (@refSeqs) {
         my $segName = $seg->{name};


This is obviously just a quick fix, and I'm not sure if it is adequate. But we will keep an eye on this to see if we can make a better fix

Thanks
-Colin

On Wed, Oct 22, 2014 at 5:42 AM, Marc Höppner <[hidden email]> wrote:
Hi,

been playing around with WebApollo and exports to a Chado db (and then onwards to a Jbrowse portal via biodb-to-json.pl ). I noticed that WebApollo does not seem to export CDS features chad, but a polypeptides instead (?). This is a bit of an issue since without CDS features the transcript is somewhat incomplete - and the CSS formatting in Jbrowse also depends on CDS features to be present in a coding transcript.

Am I missing something perhaps?

/Marc





Reply | Threaded
Open this post in threaded view
|

Re: WebApollo to chado - CDS?

Marc Höppner-3
Hi Colin,

yes, after some more digging I found the 'inferCDS' function in Bio::DB::DAS::Chado, so it is actually possible to get CDS features from Chado to JBrowse - but the documentation for biodb-to-json.pl is a bit sparse. It's a bit tricky to keep an overview with so many different bits and pieces working together (Chado, WebApollo, JBrowse). Thanks!


/Marc

Hi Marc,

Great idea. I tried testing this out and I setup a round-trip Chado export and biodb-to-json script and I can see what you mean. I don't think the polypeptide is in place of the CDS, but I found another option to get the CDS.

I found that if I added "inferCDS" to the biodb loader on jbrowse, then you can infer the CDS feaures.


diff --git a/src/perl5/Bio/JBrowse/Cmd/BioDBToJson.pm b/src/perl5/Bio/JBrowse/Cmd/BioDBToJson.pm
index d56b273..8dcfc96 100644
--- a/src/perl5/Bio/JBrowse/Cmd/BioDBToJson.pm
+++ b/src/perl5/Bio/JBrowse/Cmd/BioDBToJson.pm
@@ -74,6 +74,7 @@ sub run {
     }
     $db->strict_bounds_checking(1) if $db->can('strict_bounds_checking');
     $db->absolute(1)               if $db->can('absolute');
+    $db->inferCDS(1);
 
     foreach my $seg (@refSeqs) {
         my $segName = $seg->{name};


This is obviously just a quick fix, and I'm not sure if it is adequate. But we will keep an eye on this to see if we can make a better fix

Thanks
-Colin

On Wed, Oct 22, 2014 at 5:42 AM, Marc Höppner <[hidden email]> wrote:
Hi,

been playing around with WebApollo and exports to a Chado db (and then onwards to a Jbrowse portal via biodb-to-json.pl ). I noticed that WebApollo does not seem to export CDS features chad, but a polypeptides instead (?). This is a bit of an issue since without CDS features the transcript is somewhat incomplete - and the CSS formatting in Jbrowse also depends on CDS features to be present in a coding transcript.

Am I missing something perhaps?

/Marc






Reply | Threaded
Open this post in threaded view
|

Re: WebApollo to chado - CDS?

nathandunn
Colin is updating some of this documentation now for our next release.    

We’re actively trying to simplify how the bits work together to reduce some of these headaches.     Thanks for letting us know.

Nathan

On Oct 22, 2014, at 9:31 AM, Marc Hoeppner <[hidden email]> wrote:

Hi Colin,

yes, after some more digging I found the 'inferCDS' function in Bio::DB::DAS::Chado, so it is actually possible to get CDS features from Chado to JBrowse - but the documentation for biodb-to-json.pl is a bit sparse. It's a bit tricky to keep an overview with so many different bits and pieces working together (Chado, WebApollo, JBrowse). Thanks!


/Marc

Hi Marc,

Great idea. I tried testing this out and I setup a round-trip Chado export and biodb-to-json script and I can see what you mean. I don't think the polypeptide is in place of the CDS, but I found another option to get the CDS.

I found that if I added "inferCDS" to the biodb loader on jbrowse, then you can infer the CDS feaures.


diff --git a/src/perl5/Bio/JBrowse/Cmd/BioDBToJson.pm b/src/perl5/Bio/JBrowse/Cmd/BioDBToJson.pm
index d56b273..8dcfc96 100644
--- a/src/perl5/Bio/JBrowse/Cmd/BioDBToJson.pm
+++ b/src/perl5/Bio/JBrowse/Cmd/BioDBToJson.pm
@@ -74,6 +74,7 @@ sub run {
     }
     $db->strict_bounds_checking(1) if $db->can('strict_bounds_checking');
     $db->absolute(1)               if $db->can('absolute');
+    $db->inferCDS(1);
 
     foreach my $seg (@refSeqs) {
         my $segName = $seg->{name};


This is obviously just a quick fix, and I'm not sure if it is adequate. But we will keep an eye on this to see if we can make a better fix

Thanks
-Colin

On Wed, Oct 22, 2014 at 5:42 AM, Marc Höppner <[hidden email]> wrote:
Hi,

been playing around with WebApollo and exports to a Chado db (and then onwards to a Jbrowse portal via biodb-to-json.pl ). I noticed that WebApollo does not seem to export CDS features chad, but a polypeptides instead (?). This is a bit of an issue since without CDS features the transcript is somewhat incomplete - and the CSS formatting in Jbrowse also depends on CDS features to be present in a coding transcript.

Am I missing something perhaps?

/Marc