gmod-ajax — JBrowse list

Re: [Gmod-ajax] refSeqs & trackInfo

[Robert B. <rb...@gm...>]

Hi Jerome,

In JBrowse 1.2.1, the data/refSeqs.js and data/trackInfo.js must be 
included in the index.html.

In the coming release, however, this will be passable as a parameter to 
the JBrowse object.

Rob

On 03/01/2012 11:28 AM, Mariette wrote:
> Hi everyone,
>
> I have hard time to understand how is managed the refSeqs&  the
> trackInfo params when building a jbrowse object.
> I'd like to set the path to the js but it's not working. There is any
> way to provide the right path to the js file other than
> including it inside the html page ?
>
> thanks for your reply,
> Jerome
>
> ------------------------------------------------------------------------------
> Virtualization&  Cloud Management Using Capacity Planning
> Cloud computing makes use of virtualization - but cloud computing
> also focuses on allowing computing to be delivered as a service.
> http://www.accelacomm.com/jaw/sfnl/114/51521223/
> _______________________________________________
> Gmod-ajax mailing list
> Gmo...@li...
> https://lists.sourceforge.net/lists/listinfo/gmod-ajax
>

[Gmod-ajax] refSeqs & trackInfo

[Mariette <jma...@to...>]

Hi everyone,

I have hard time to understand how is managed the refSeqs & the 
trackInfo params when building a jbrowse object.
I'd like to set the path to the js but it's not working. There is any 
way to provide the right path to the js file other than
including it inside the html page ?

thanks for your reply,
Jerome

Re: [Gmod-ajax] setting up jbrowse with a bacterial reference genome

[Heewook L. <he...@um...>]

0-based half-open explains but then the bug is not the interval but
rendering/drawing of elements because if you have an element that is
single-base event such as SNP, the drawing will take up 2bp rather than
1bp.

For example, SNP is at 1,000 in genbank file(which is 1-based), then when
you add this to jbrowse, this will be displayed as (Start: 999, end:1000).
Since the end interval is open interval 1000 should not be included in the
drawing, however, if you zoom in all the way you can surely tell that
jbrowse will render this event as 2bp.

Right now, i am using triangle from genome.css to display a SNP and when
you zoom all the way in(1bp-resolution), the triangle will turn
into isosceles trapezoid because it takes up 2bp.

-hl

On Wed, Feb 29, 2012 at 5:21 PM, Karsten Hokamp <kh...@gm...> wrote:

> Hi,
>
> I'm not sure this was a bug. In a previous discussion it was noted
> that JBrowse was 0-based, see from message 5 onwards in this thread:
>
> http://gmod.827538.n3.nabble.com/Gmod-ajax-Formatting-using-GFF-notes-and-some-other-stuff-td998491.html
>
> However, there was mentioning of changing this to 1-based for the user
> interface and I suppose this is done now.
>
> Thanks,
>
>    Karsten
>
>
> On Wed, Feb 29, 2012 at 8:06 PM, Robert Buels <rb...@gm...> wrote:
> > Fixed in master, issue URL: https://github.com/gmod/jbrowse/issues/73
> >
> > On 02/28/2012 02:11 PM, Heewook Lee wrote:
> >> Done!
> >>
> >> On Tue, Feb 28, 2012 at 1:52 PM, Robert Buels <rb...@gm...
> >> <mailto:rb...@gm...>> wrote:
> >>
> >>     Hi Heewook,
> >>
> >>     I'll take a look.  Could you add this to the JBrowse issue tracker
> at
> >>     https://github.com/GMOD/__jbrowse/issues
> >>     <https://github.com/GMOD/jbrowse/issues>?
> >>
> >>     Rob
> >>
> >>
> >>     On 02/27/2012 01:17 PM, Heewook Lee wrote:
> >>
> >>         I have noticed all entries in gff when it's added to jbrowse via
> >>         biodb-to-json.pl <http://biodb-to-json.pl>
> >>         <http://biodb-to-json.pl/>, start index  always
> >>
> >>         gets offset by 1. For example, if you have an entry in gff file
> and
> >>         start and end is 1000 2000, when it's added on jbrowse, it will
> >>         display
> >>         as "start: 999, end: 2000. Why does this happen and are you
> aware of
> >>         this? Thanks!
> >>
> >>
> >
> >
> >
> ------------------------------------------------------------------------------
> > Virtualization & Cloud Management Using Capacity Planning
> > Cloud computing makes use of virtualization - but cloud computing
> > also focuses on allowing computing to be delivered as a service.
> > http://www.accelacomm.com/jaw/sfnl/114/51521223/
> > _______________________________________________
> > Gmod-ajax mailing list
> > Gmo...@li...
> > https://lists.sourceforge.net/lists/listinfo/gmod-ajax
>

Re: [Gmod-ajax] setting up jbrowse with a bacterial reference genome

[Robert B. <rb...@gm...>]

Ah, I'd missed this up to now.  Yes, it seems it's all 0-based 
half-open.  I'll revert that commit.  Thanks a lot for pointing that out 
Karsten.

I'll add a ticket to convert everything shown to the user to be 1-based 
closed coordinates, because that's what people expect to see.

Rob

On 02/29/2012 05:21 PM, Karsten Hokamp wrote:
> Hi,
>
> I'm not sure this was a bug. In a previous discussion it was noted
> that JBrowse was 0-based, see from message 5 onwards in this thread:
> http://gmod.827538.n3.nabble.com/Gmod-ajax-Formatting-using-GFF-notes-and-some-other-stuff-td998491.html
>
> However, there was mentioning of changing this to 1-based for the user
> interface and I suppose this is done now.
>
> Thanks,
>
>      Karsten
>
>
> On Wed, Feb 29, 2012 at 8:06 PM, Robert Buels<rb...@gm...>  wrote:
>> Fixed in master, issue URL: https://github.com/gmod/jbrowse/issues/73
>>
>> On 02/28/2012 02:11 PM, Heewook Lee wrote:
>>> Done!
>>>
>>> On Tue, Feb 28, 2012 at 1:52 PM, Robert Buels<rb...@gm...
>>> <mailto:rb...@gm...>>  wrote:
>>>
>>>      Hi Heewook,
>>>
>>>      I'll take a look.  Could you add this to the JBrowse issue tracker at
>>>      https://github.com/GMOD/__jbrowse/issues
>>>      <https://github.com/GMOD/jbrowse/issues>?
>>>
>>>      Rob
>>>
>>>
>>>      On 02/27/2012 01:17 PM, Heewook Lee wrote:
>>>
>>>          I have noticed all entries in gff when it's added to jbrowse via
>>>          biodb-to-json.pl<http://biodb-to-json.pl>
>>>          <http://biodb-to-json.pl/>, start index  always
>>>
>>>          gets offset by 1. For example, if you have an entry in gff file and
>>>          start and end is 1000 2000, when it's added on jbrowse, it will
>>>          display
>>>          as "start: 999, end: 2000. Why does this happen and are you aware of
>>>          this? Thanks!
>>>
>>>
>>
>>
>> ------------------------------------------------------------------------------
>> Virtualization&  Cloud Management Using Capacity Planning
>> Cloud computing makes use of virtualization - but cloud computing
>> also focuses on allowing computing to be delivered as a service.
>> http://www.accelacomm.com/jaw/sfnl/114/51521223/
>> _______________________________________________
>> Gmod-ajax mailing list
>> Gmo...@li...
>> https://lists.sourceforge.net/lists/listinfo/gmod-ajax
>

Re: [Gmod-ajax] setting up jbrowse with a bacterial reference genome

[Karsten H. <kh...@gm...>]

Hi,

I'm not sure this was a bug. In a previous discussion it was noted
that JBrowse was 0-based, see from message 5 onwards in this thread:
http://gmod.827538.n3.nabble.com/Gmod-ajax-Formatting-using-GFF-notes-and-some-other-stuff-td998491.html

However, there was mentioning of changing this to 1-based for the user
interface and I suppose this is done now.

Thanks,

    Karsten


On Wed, Feb 29, 2012 at 8:06 PM, Robert Buels <rb...@gm...> wrote:
> Fixed in master, issue URL: https://github.com/gmod/jbrowse/issues/73
>
> On 02/28/2012 02:11 PM, Heewook Lee wrote:
>> Done!
>>
>> On Tue, Feb 28, 2012 at 1:52 PM, Robert Buels <rb...@gm...
>> <mailto:rb...@gm...>> wrote:
>>
>>     Hi Heewook,
>>
>>     I'll take a look.  Could you add this to the JBrowse issue tracker at
>>     https://github.com/GMOD/__jbrowse/issues
>>     <https://github.com/GMOD/jbrowse/issues>?
>>
>>     Rob
>>
>>
>>     On 02/27/2012 01:17 PM, Heewook Lee wrote:
>>
>>         I have noticed all entries in gff when it's added to jbrowse via
>>         biodb-to-json.pl <http://biodb-to-json.pl>
>>         <http://biodb-to-json.pl/>, start index  always
>>
>>         gets offset by 1. For example, if you have an entry in gff file and
>>         start and end is 1000 2000, when it's added on jbrowse, it will
>>         display
>>         as "start: 999, end: 2000. Why does this happen and are you aware of
>>         this? Thanks!
>>
>>
>
>
> ------------------------------------------------------------------------------
> Virtualization & Cloud Management Using Capacity Planning
> Cloud computing makes use of virtualization - but cloud computing
> also focuses on allowing computing to be delivered as a service.
> http://www.accelacomm.com/jaw/sfnl/114/51521223/
> _______________________________________________
> Gmod-ajax mailing list
> Gmo...@li...
> https://lists.sourceforge.net/lists/listinfo/gmod-ajax

Re: [Gmod-ajax] setting up jbrowse with a bacterial reference genome

[Heewook L. <he...@um...>]

Thank you so much! just surprised this went undetected till now. :)

-hl

On Wed, Feb 29, 2012 at 3:06 PM, Robert Buels <rb...@gm...> wrote:

> Fixed in master, issue URL: https://github.com/gmod/**jbrowse/issues/73<https://github.com/gmod/jbrowse/issues/73>
>
>
> On 02/28/2012 02:11 PM, Heewook Lee wrote:
>
>> Done!
>>
>> On Tue, Feb 28, 2012 at 1:52 PM, Robert Buels <rb...@gm...
>> <mailto:rb...@gm...>> wrote:
>>
>>    Hi Heewook,
>>
>>    I'll take a look.  Could you add this to the JBrowse issue tracker at
>>    https://github.com/GMOD/__**jbrowse/issues<https://github.com/GMOD/__jbrowse/issues>
>>    <https://github.com/GMOD/**jbrowse/issues<https://github.com/GMOD/jbrowse/issues>
>> >?
>>
>>
>>    Rob
>>
>>
>>    On 02/27/2012 01:17 PM, Heewook Lee wrote:
>>
>>        I have noticed all entries in gff when it's added to jbrowse via
>>        biodb-to-json.pl <http://biodb-to-json.pl>
>>        <http://biodb-to-json.pl/>, start index  always
>>
>>        gets offset by 1. For example, if you have an entry in gff file and
>>        start and end is 1000 2000, when it's added on jbrowse, it will
>>        display
>>        as "start: 999, end: 2000. Why does this happen and are you aware
>> of
>>        this? Thanks!
>>
>>
>>
>

Re: [Gmod-ajax] setting up jbrowse with a bacterial reference genome

[Robert B. <rb...@gm...>]

Fixed in master, issue URL: https://github.com/gmod/jbrowse/issues/73

On 02/28/2012 02:11 PM, Heewook Lee wrote:
> Done!
>
> On Tue, Feb 28, 2012 at 1:52 PM, Robert Buels <rb...@gm...
> <mailto:rb...@gm...>> wrote:
>
>     Hi Heewook,
>
>     I'll take a look.  Could you add this to the JBrowse issue tracker at
>     https://github.com/GMOD/__jbrowse/issues
>     <https://github.com/GMOD/jbrowse/issues>?
>
>     Rob
>
>
>     On 02/27/2012 01:17 PM, Heewook Lee wrote:
>
>         I have noticed all entries in gff when it's added to jbrowse via
>         biodb-to-json.pl <http://biodb-to-json.pl>
>         <http://biodb-to-json.pl/>, start index  always
>
>         gets offset by 1. For example, if you have an entry in gff file and
>         start and end is 1000 2000, when it's added on jbrowse, it will
>         display
>         as "start: 999, end: 2000. Why does this happen and are you aware of
>         this? Thanks!
>
>

Re: [Gmod-ajax] Problems when compiling wig2png of JBrowse

[王聃 <wan...@16...>]

OK,Rob.
At 2012-02-28 22:57:53,"Robert Buels" <rb...@gm...> wrote:
>Hi Dan,
>
>Could you re-send that reply to the mailing list, so there's a public 
>record of how you solved it?
>
>Rob
>
>On 02/27/2012 11:05 PM, 王聃 wrote:
>> Thanks Rob!
>> I've solved the problem now and maybe something wrong with g++-4.6. I
>> used g++-4.4 instead and compiled it successfully somehow.
>> I'm honored to be a member of jbrowse and Thanks for your help.
>> Dan
>>
>> At  2012-02-27  23:52:38,"Robert  Buels"  <rb...@gm...>  wrote:
>>>Hi  Dan,
>>>
>>>Do  you  have  the  libpng  development  headers  installed?    To  make  sure,  try
>>>"sudo  apt-get  install  libpng15-dev",  then  run  ./configure  and  make  again.
>>>
>>>Rob
>>>
>>>On  02/27/2012  02:27  AM,  wangdan  wrote:
>>>>  Hello,  everyone!
>>>>  I've  installed  libpng12-0  and  libpng12-dev  on  my  OS  ubuntu  11.10
>>>>  i386.There's  something  wrong  when  I  compiling  wig2png.  I've  run
>>>>  configure  first  and  I've  installed  g++  4.6  and  make  tools  using  "sudo".
>>>>  It  showed  the  errors  in  the  terminal:
>>>>  /tmp/ccNbdLZq.o:  In  function  `write_png_file(char  const*,  int,  int,
>>>>  unsigned  char**,  png_color_struct*,  int,  int)':
>>>>  /var/www/jbrowse-1.2.1/src/wig2png.cc:47:  undefined  reference  to
>>>>  `png_create_write_struct'
>>>snip
>>>>
>>>>  Actually  I've  tried  different  version  of  libpng  instead.  I  compiled  and
>>>>  installed  libpng15  and  tried  wig2png  again  but  it  didn't  work  yet.
>>>>  What's  the  problem?  How  to  solve  this?
>>>>  Thank  you  very  much!
>>>>  Wang  Dan
>>
>>
>>
>

Re: [Gmod-ajax] Displaying all of the attributes in gff3 file?

[Robert B. <rb...@gm...>]

Hi Heewook,

The current JBrowse release (version 1.2.1) does not have a way to do this.

This is a major shortcoming, and I've added it to the issue tracker: 
https://github.com/GMOD/jbrowse/issues/72

Rob


On 02/27/2012 01:38 PM, Heewook Lee wrote:
> Hi,
>
> I am using biodb-to-json.pl <http://biodb-to-json.pl> with a gff3 file
> and I noticed that all the attributes on my gff file (except for Name
> and Alias) are being ignored) Is there way to display these attributes
> upon clicking the entry shown on jBrowse? Thanks
>
> -hl

Re: [Gmod-ajax] setting up jbrowse with a bacterial reference genome

[Heewook L. <he...@um...>]

Done!

On Tue, Feb 28, 2012 at 1:52 PM, Robert Buels <rb...@gm...> wrote:

> Hi Heewook,
>
> I'll take a look.  Could you add this to the JBrowse issue tracker at
> https://github.com/GMOD/**jbrowse/issues<https://github.com/GMOD/jbrowse/issues>
> ?
>
> Rob
>
>
> On 02/27/2012 01:17 PM, Heewook Lee wrote:
>
>> I have noticed all entries in gff when it's added to jbrowse via
>> biodb-to-json.pl <http://biodb-to-json.pl/>, start index  always
>>
>> gets offset by 1. For example, if you have an entry in gff file and
>> start and end is 1000 2000, when it's added on jbrowse, it will display
>> as "start: 999, end: 2000. Why does this happen and are you aware of
>> this? Thanks!
>>
>

Re: [Gmod-ajax] setting up jbrowse with a bacterial reference genome

[Robert B. <rb...@gm...>]

Hi Heewook,

I'll take a look.  Could you add this to the JBrowse issue tracker at
https://github.com/GMOD/jbrowse/issues?

Rob

On 02/27/2012 01:17 PM, Heewook Lee wrote:
> I have noticed all entries in gff when it's added to jbrowse via
> biodb-to-json.pl <http://biodb-to-json.pl/>, start index  always
> gets offset by 1. For example, if you have an entry in gff file and
> start and end is 1000 2000, when it's added on jbrowse, it will display
> as "start: 999, end: 2000. Why does this happen and are you aware of
> this? Thanks!

[Gmod-ajax] Show RC, highlight regions

[Ian D. <ian...@gr...>]

Hi JBrowse developers,

I'm working on a little project to mashup JBrowse with BLAST, so I can see where my BLAST hits fall in the genome.  So far, so good - I have JBrowse in an IFRAME and <a href=... target='jbrowse'> type links in my BLAST results.

Two minor feature requests, though.  First , I'd really like the sequence view to show both strands, rather than just one - both strands was the default for GBrowse.  Second, I'd like to be able to highlight regions of the DNA sequence, e.g., the region involved in my BLAST hit.  (In fact, more features to support "dynamic" annotation of JBrowse objects using JavaScript, on the fly, would be most welcome :)

So, I've implemented both these features and am attaching a patch.  The highlighting is quite hacky - it just highlights the initial position of the browser, and hangs on to that forever.  It should probably be a separate URL parameter, but I was having enough trouble as-is figuring out how to make it work.  Maybe someone on the team could suggest a more elegant solution?  The reverse complement is ~6 lines of code - just needed to add a <BR> and the sequence-complement function.  Hopefully that can be used as is.  (I believe RC is already a feature request, #20 on GitHub.)

I should say, much thanks for creating JBrowse.  I think it's a big step forward overall, compared to GBrowse.  I'll be interested to hear your thoughts on this patch - I might be interested in implementing a few of the other missing features :)

Best,
Ian

---
Ian W. Davis, PhD
Head of Computing / Bioinformatics
GrassRoots Biotechnology


diff --git a/js/Browser.js b/js/Browser.js
index 6ccd17d..3239476 100644
--- a/js/Browser.js
+++ b/js/Browser.js
@@ -158,6 +158,21 @@ var Browser = function(params) {
                   for (var i = 0; i < brwsr.deferredFunctions.length; i++)
                               brwsr.deferredFunctions[i]();
                   brwsr.deferredFunctions = [];
+
+            // Capture initial location for highlighting DNA sequence.
+            // Tracks are initialized by one of the deferred functions!
+            var hilightLoc = { refName: brwsr.view.ref.name,
+                                refStart: Math.round(brwsr.view.minVisible()),
+                                refEnd: Math.round(brwsr.view.maxVisible()) };
+            for(var i = 0; i < brwsr.view.tracks.length; i++) {
+                var track = brwsr.view.tracks[i];
+                if(track.name == "DNA") {
+                    track.hilightLoc = hilightLoc;
+                    track.clear();
+                }
+            }
+            brwsr.view.showVisibleBlocks(true);
+            // Calls to .clear() and .showVisibleBlocks() required to refresh the view...
         });
};
diff --git a/js/SequenceTrack.js b/js/SequenceTrack.js
index 32a1818..aa06337 100644
--- a/js/SequenceTrack.js
+++ b/js/SequenceTrack.js
@@ -19,11 +19,13 @@ function SequenceTrack(trackMeta, url, refSeq, browserParams) {
     Track.call(this, trackMeta.label, trackMeta.key,
                false, browserParams.changeCallback);
     this.browserParams = browserParams;
+    this.refSeq = refSeq;
     this.trackMeta = trackMeta;
     this.setLoaded();
     this.chunks = [];
    this.chunkSize = trackMeta.args.chunkSize;
     this.baseUrl = (browserParams.baseUrl ? browserParams.baseUrl : "") + url;
+    this.hilightLoc = {refName: null, refStart:-1, refEnd:-1};
}
 SequenceTrack.prototype = new Track("");
@@ -57,13 +59,38 @@ SequenceTrack.prototype.fillBlock = function(blockIndex, block,
                                              leftBase, rightBase,
                                              scale, stripeWidth,
                                              containerStart, containerEnd) {
+    var compl_rx = /[ACGT]/g;
+    var compl_tbl = {'A':'G', 'C':'T', 'G':'A', 'T':'C'};
+    var compl_fn = function(m) { return compl_tbl[m] || m; }
+    var hloc = this.hilightLoc;
+    var rseq = this.refSeq;
+    function hilighted_seq(start, end, seq) {
+        // start, end coords are interbase (half-open intervals)
+        if(hloc.refName == rseq.name & hloc.refStart < end & hloc.refEnd > start) {
+            var hseq_start = Math.max(0, hloc.refStart - start);
+            var hseq_end = Math.min(seq.length, hloc.refEnd - start);
+            var spanOuter = document.createElement("span");
+            var spanInner = document.createElement("span");
+            spanInner.style.cssText = "background: #ff0;";
+            spanOuter.appendChild(document.createTextNode(seq.substring(0, hseq_start)));
+            spanInner.appendChild(document.createTextNode(seq.substring(hseq_start, hseq_end)));
+            spanOuter.appendChild(spanInner);
+            spanOuter.appendChild(document.createTextNode(seq.substring(hseq_end, seq.length)));
+            return spanOuter;
+        } else {
+            return document.createTextNode(seq);
+        }
+    }
     if (this.shown) {
         this.getRange(leftBase, rightBase,
                       function(start, end, seq) {
                           //console.log("adding seq from %d to %d: %s", start, end, seq);
                           var seqNode = document.createElement("div");
                           seqNode.className = "sequence";
-                          seqNode.appendChild(document.createTextNode(seq));
+                          seqNode.appendChild(hilighted_seq(start, end, seq));
+                          // IWD: add complement!
+                          seqNode.appendChild(document.createElement("br"));
+                          seqNode.appendChild(hilighted_seq(start, end, seq.replace(compl_rx, compl_fn)));
                                 seqNode.style.cssText = "top: 0px;";
                           block.appendChild(seqNode);
                       });

[Gmod-ajax] Displaying all of the attributes in gff3 file?

[Heewook L. <he...@um...>]

Hi,

I am using biodb-to-json.pl with a gff3 file and I noticed that all the
attributes on my gff file (except for Name and Alias) are being ignored) Is
there way to display these attributes upon clicking the entry shown on
jBrowse? Thanks

-hl

Re: [Gmod-ajax] setting up jbrowse with a bacterial reference genome

[Heewook L. <he...@um...>]

Rob,

Actually i got this solved somehow.. instead of adding single track at a
time with the flatfile script, I tured to biodb-to-json.pl. Anyhow, I have
one more question:

I have noticed all entries in gff when it's added to jbrowse via
biodb-to-json.pl, start index  always gets offset by 1. For example, if you
have an entry in gff file and start and end is 1000 2000, when it's added
on jbrowse, it will display as "start: 999, end: 2000. Why does this happen
and are you aware of this? Thanks!

On Mon, Feb 27, 2012 at 11:29 AM, Robert Buels <rb...@gm...> wrote:

> Hi Heewook,
>
> Do you have the browser online anywhere that I can look at it?  That would
> help me diagnose what's going on.
>
> If you can't put it online, could you email me the
> data/tracks/*/SNP/trackData.**json and data/trackInfo.js files so I can
> look at them?
>
> Rob
>
>
>
> On 02/25/2012 01:44 AM, Heewook Lee wrote:
>
>> Thanks. It works great!
>>
>> I have one more question though. So now I have set up the jbrowse (using
>> gbk2gff then biodb-to-json.pl <http://biodb-to-json.pl> with a config
>>
>> file similar to yeast configuration file you provide with the tarball)
>> and now I want to add a track for displaying point mutations. I have
>> already prepared the point mutations in a gff3 format so each point
>> mutation in gff file is as :
>>
>> ecoliSNP90109010...ID=ecoli_**9010;Name=ecoli_9010;**
>> Reference_Base=C;Consensus=C;**Modified_Call=T;snp_type=N-Syn
>>
>>
>> since i have these point mutations in a separate file, I
>> used bin/flatfile-to-json.pl <http://flatfile-to-json.pl> :
>>
>> $bin/flatfile-to-json.pl <http://flatfile-to-json.pl> --gff
>>
>> pointMutations.gff --tracklabel SNP --key SNP
>>
>> After this, when I refresh the jbrowse, I can see that new track, SNP
>> has been added but the problem is it's not being rendered on the
>> browser. When I zoom out enough I can see a histogram for the track but
>> when i zoom in, I see nothing. What would I need to do to display SNPs
>> correctly? Thanks!
>>
>>
>>
>> On Wed, Feb 15, 2012 at 12:14 PM, Robert Buels <rb...@gm...
>> <mailto:rb...@gm...>> wrote:
>>
>>    Hi Heewook,
>>
>>    Bio::DB::SeqFeature::Store's behavior has changed slightly since
>>    that Current Protocols publication was written.  In the
>>    NC_001133.gbk.gff, just remove the ##sequence-region line that the
>>    directions in the publication told you to alter, and the error
>>    should go away.  Sorry for the inconvenience.
>>
>>    Rob
>>
>>
>>
>>    On 02/10/2012 01:32 PM, Heewook Lee wrote:
>>
>>        -------------------- EXCEPTION --------------------
>>        MSG: segment() called in a scalar context but multiple features
>>        match.
>>        Either call in a list context or narrow your search using the
>>        -types or
>>        -class arguments
>>
>>        STACK Bio::DB::SeqFeature::Store::__**segment
>>        /usr/local/share/perl5/Bio/DB/**__SeqFeature/Store.pm:1359
>>        STACK toplevel bin/prepare-refseqs.pl:156
>>        <http://prepare-refseqs.pl:156**> <http://prepare-refseqs.pl:**
>> 156__>
>>
>>        ------------------------------**__-------------
>>
>>
>>        Any idea how I can get around this? Thanks
>>
>>
>>
>

Re: [Gmod-ajax] setting up jbrowse with a bacterial reference genome

[Robert B. <rb...@gm...>]

Hi Heewook,

Do you have the browser online anywhere that I can look at it?  That 
would help me diagnose what's going on.

If you can't put it online, could you email me the 
data/tracks/*/SNP/trackData.json and data/trackInfo.js files so I can 
look at them?

Rob


On 02/25/2012 01:44 AM, Heewook Lee wrote:
> Thanks. It works great!
>
> I have one more question though. So now I have set up the jbrowse (using
> gbk2gff then biodb-to-json.pl <http://biodb-to-json.pl> with a config
> file similar to yeast configuration file you provide with the tarball)
> and now I want to add a track for displaying point mutations. I have
> already prepared the point mutations in a gff3 format so each point
> mutation in gff file is as :
>
> ecoliSNP90109010...ID=ecoli_9010;Name=ecoli_9010;Reference_Base=C;Consensus=C;Modified_Call=T;snp_type=N-Syn
>
> since i have these point mutations in a separate file, I
> used bin/flatfile-to-json.pl <http://flatfile-to-json.pl> :
>
> $bin/flatfile-to-json.pl <http://flatfile-to-json.pl> --gff
> pointMutations.gff --tracklabel SNP --key SNP
>
> After this, when I refresh the jbrowse, I can see that new track, SNP
> has been added but the problem is it's not being rendered on the
> browser. When I zoom out enough I can see a histogram for the track but
> when i zoom in, I see nothing. What would I need to do to display SNPs
> correctly? Thanks!
>
>
>
> On Wed, Feb 15, 2012 at 12:14 PM, Robert Buels <rb...@gm...
> <mailto:rb...@gm...>> wrote:
>
>     Hi Heewook,
>
>     Bio::DB::SeqFeature::Store's behavior has changed slightly since
>     that Current Protocols publication was written.  In the
>     NC_001133.gbk.gff, just remove the ##sequence-region line that the
>     directions in the publication told you to alter, and the error
>     should go away.  Sorry for the inconvenience.
>
>     Rob
>
>
>
>     On 02/10/2012 01:32 PM, Heewook Lee wrote:
>
>         -------------------- EXCEPTION --------------------
>         MSG: segment() called in a scalar context but multiple features
>         match.
>         Either call in a list context or narrow your search using the
>         -types or
>         -class arguments
>
>         STACK Bio::DB::SeqFeature::Store::__segment
>         /usr/local/share/perl5/Bio/DB/__SeqFeature/Store.pm:1359
>         STACK toplevel bin/prepare-refseqs.pl:156
>         <http://prepare-refseqs.pl:156> <http://prepare-refseqs.pl:156__>
>
>         ------------------------------__-------------
>
>         Any idea how I can get around this? Thanks
>
>

Re: [Gmod-ajax] Problems when compiling wig2png of JBrowse

[Robert B. <rb...@gm...>]

Hi Dan,

Do you have the libpng development headers installed?  To make sure, try 
"sudo apt-get install libpng15-dev", then run ./configure and make again.

Rob

On 02/27/2012 02:27 AM, wangdan wrote:
> Hello, everyone!
> I've installed libpng12-0 and libpng12-dev on my OS ubuntu 11.10
> i386.There's something wrong when I compiling wig2png. I've run
> configure first and I've installed g++ 4.6 and make tools using "sudo".
> It showed the errors in the terminal:
> /tmp/ccNbdLZq.o: In function `write_png_file(char const*, int, int,
> unsigned char**, png_color_struct*, int, int)':
> /var/www/jbrowse-1.2.1/src/wig2png.cc:47: undefined reference to
> `png_create_write_struct'
snip
>
> Actually I've tried different version of libpng instead. I compiled and
> installed libpng15 and tried wig2png again but it didn't work yet.
> What's the problem? How to solve this?
> Thank you very much!
> Wang Dan

[Gmod-ajax] Problems when compiling wig2png of JBrowse

[wangdan <wan...@16...>]

Hello, everyone!
I've installed libpng12-0 and libpng12-dev on my OS ubuntu 11.10
i386.There's something wrong when I compiling wig2png. I've run
configure first and I've installed g++ 4.6 and make tools using "sudo".
It showed the errors in the terminal:
/tmp/ccNbdLZq.o: In function `write_png_file(char const*, int, int,
unsigned char**, png_color_struct*, int, int)':
/var/www/jbrowse-1.2.1/src/wig2png.cc:47: undefined reference to
`png_create_write_struct'
/var/www/jbrowse-1.2.1/src/wig2png.cc:53: undefined reference to
`png_create_info_struct'
/var/www/jbrowse-1.2.1/src/wig2png.cc:60: undefined reference to
`png_set_longjmp_fn'
/var/www/jbrowse-1.2.1/src/wig2png.cc:66: undefined reference to
`png_init_io'
/var/www/jbrowse-1.2.1/src/wig2png.cc:71: undefined reference to
`png_set_IHDR'
/var/www/jbrowse-1.2.1/src/wig2png.cc:74: undefined reference to
`png_set_PLTE'
/var/www/jbrowse-1.2.1/src/wig2png.cc:77: undefined reference to
`png_set_rows'
/var/www/jbrowse-1.2.1/src/wig2png.cc:81: undefined reference to
`png_set_filter'
/var/www/jbrowse-1.2.1/src/wig2png.cc:82: undefined reference to
`png_set_compression_level'
/var/www/jbrowse-1.2.1/src/wig2png.cc:86: undefined reference to
`png_write_png'
/var/www/jbrowse-1.2.1/src/wig2png.cc:88: undefined reference to
`png_destroy_write_struct'
/var/www/jbrowse-1.2.1/src/wig2png.cc:61: undefined reference to
`png_destroy_write_struct'
/var/www/jbrowse-1.2.1/src/wig2png.cc:55: undefined reference to
`png_destroy_write_struct'

Actually I've tried different version of libpng instead. I compiled and
installed libpng15 and tried wig2png again but it didn't work yet.
What's the problem? How to solve this?
Thank you very much!
Wang Dan

Re: [Gmod-ajax] setting up jbrowse with a bacterial reference genome

[Heewook L. <he...@um...>]

Thanks. It works great!

I have one more question though. So now I have set up the jbrowse (using
gbk2gff then biodb-to-json.pl with a config file similar to yeast
configuration file you provide with the tarball) and now I want to add a
track for displaying point mutations. I have already prepared the point
mutations in a gff3 format so each point mutation in gff file is as :

ecoli SNP 9010 9010 . . .
ID=ecoli_9010;Name=ecoli_9010;Reference_Base=C;Consensus=C;Modified_Call=T;snp_type=N-Syn

since i have these point mutations in a separate file, I used bin/
flatfile-to-json.pl :

$bin/flatfile-to-json.pl --gff pointMutations.gff --tracklabel SNP --key SNP

After this, when I refresh the jbrowse, I can see that new track, SNP has
been added but the problem is it's not being rendered on the browser. When
I zoom out enough I can see a histogram for the track but when i zoom in, I
see nothing. What would I need to do to display SNPs correctly? Thanks!



On Wed, Feb 15, 2012 at 12:14 PM, Robert Buels <rb...@gm...> wrote:

> Hi Heewook,
>
> Bio::DB::SeqFeature::Store's behavior has changed slightly since that
> Current Protocols publication was written.  In the NC_001133.gbk.gff, just
> remove the ##sequence-region line that the directions in the publication
> told you to alter, and the error should go away.  Sorry for the
> inconvenience.
>
> Rob
>
>
>
> On 02/10/2012 01:32 PM, Heewook Lee wrote:
>
>> -------------------- EXCEPTION --------------------
>> MSG: segment() called in a scalar context but multiple features match.
>> Either call in a list context or narrow your search using the -types or
>> -class arguments
>>
>> STACK Bio::DB::SeqFeature::Store::**segment
>> /usr/local/share/perl5/Bio/DB/**SeqFeature/Store.pm:1359
>> STACK toplevel bin/prepare-refseqs.pl:156 <http://prepare-refseqs.pl:156*
>> *>
>>
>> ------------------------------**-------------
>>
>> Any idea how I can get around this? Thanks
>>
>

Re: [Gmod-ajax] setting up jbrowse with a bacterial reference genome

[Robert B. <rb...@gm...>]

Hi Heewook,

Bio::DB::SeqFeature::Store's behavior has changed slightly since that 
Current Protocols publication was written.  In the NC_001133.gbk.gff, 
just remove the ##sequence-region line that the directions in the 
publication told you to alter, and the error should go away.  Sorry for 
the inconvenience.

Rob


On 02/10/2012 01:32 PM, Heewook Lee wrote:
> -------------------- EXCEPTION --------------------
> MSG: segment() called in a scalar context but multiple features match.
> Either call in a list context or narrow your search using the -types or
> -class arguments
>
> STACK Bio::DB::SeqFeature::Store::segment
> /usr/local/share/perl5/Bio/DB/SeqFeature/Store.pm:1359
> STACK toplevel bin/prepare-refseqs.pl:156 <http://prepare-refseqs.pl:156>
> -------------------------------------------
>
> Any idea how I can get around this? Thanks

[Gmod-ajax] JBrowse meeting minutes - Feb 13, 2012

[Robert B. <rb...@gm...>]

Attendees
=========
Ed Lee
Gregg Helt
Nomi Harris
Chris Childers
Rob Buels

Minutes
============

Since the last meeting, Rob has been working steadily to finish up the
json-stabilize branch, which, predictably, is requiring much more work 
than he thought.

== JSON-stabilize (i.e. JSON format improvements) ==

Some discussion about non-backward-compatibility of the old versus new
data formats:

Chris C: each of his JBrowse installations is basically its own
sandbox. If there are significant improvments, would be good to have a
way to migrate installations to the new format.

Rob: there will be signif. improvements after Natalie Fox's faceted
browsing is folded in (it turned out to have issues that made it hard
to integrate--track labels are conflated with track IDs, etc.).

Gregg: for us, a conversion script (old -> new data format) isn't a
high priority, but it might matter more to other users. We're planning
to generate new processed data for honeybee anyway, so we can just use
the new format for that.

Rob: OK, so a migration script will be a higher priority when there
are significant UI improvements that make a compelling case for users
to upgrade existing installations.

Gregg has been experimenting with some client-side code based on work
by Thomas Down (of Dalliance) for reading remotely-hosted BAM files
using HTTP byte-range requests.  One down-side is that the entire BAM
index must still be downloaded to the the client first, and these can
run more than 1 MB.  Rob pointed out that these will usually be
compressed greatly during transfer if the web server is doing
compression, which should mitigate this somewhat.

One sticking point right now is that the nested containment list data
structure used by JBrowse works well for most data but is awkward for
directly reading BAM files.  Building nested containment lists on the
fly in the client may be very inefficient. This is making Gregg
consider replacing nested containment lists as the underlying JBrowse
data structure. He'd like to generalize the code so it's calling
abstract methods on the underlying data structure so that it could be
something else.  Rob agrees that it would be great to get some
abstraction into the data representation.  [Stuff about nested
containment lists and data structures]


== Differences Between Old and New JSON formats ==

Rob has produced four GitHub Gists showing differences between the old
and new JSON formats:

https://gist.github.com/1820831
https://gist.github.com/1820875
https://gist.github.com/1820918
https://gist.github.com/1820969

== Julie Moon ==

Intern Julie Moon at OICR is doing some good work on JBrowse.  She has
implemented an 'embedded' running mode for JBrowse that can be used to
show a certain genome region, leaving out the track listing, overview
panel, and navigation buttons, and adding a link to a 'full view' URL.
Next, she will be working on replacing the current selection box for
reference sequences with an autocompletion ComboBox that should be
much easier to work with for large numbers of reference sequences.

== Rob visit to Berkeley ==

Rob B. will be in town March 18-27. He suggests having the next
JBrowse meeting on March 12 (usual time, 2pm PST), in order to plan
activities in advance of his visit. He will check with Ian about the
date.

== Next Meeting ==

   March 12, 2pm Pacific / 5pm Eastern
   (need to confirm that time with Ian)

[Gmod-ajax] Reminder: JBrowse conference call, February 13th, 2pm Pacific

[Robert B. <rb...@gm...>]

Hello JBrowse users and developers,

As a reminder, the next JBrowse developers' conference call will be held 
on Monday, February 13, at 2pm Pacific time, which is in about 20 
minutes.  Call details are below; we hope you will can oin in.

Call details:

1. Dial Toll-Free Number: 866-740-1260 (U.S. & Canada)
(The toll-free number from Thailand is 001800656911)
2. Enter access code 4867508#

Talk to you then!

Rob
----
Robert Buels
JBrowse Lead Developer

Re: [Gmod-ajax] setting up jbrowse with a bacterial reference genome

[Heewook L. <he...@um...>]

I am actually looking at a document: Setting Up the JBrowse Genome Browser
that was published in Current Protocols in Bioinformatics in 2010 and there
is an example using NCBI gbk file. I tried that but I am getting the
following error:

[heewlee@ngs tmp]$ bin/prepare-refseqs.pl --conf
docs/examples/config/yeast_genbank.json --refs NC_001133

-------------------- EXCEPTION --------------------
MSG: segment() called in a scalar context but multiple features match.
Either call in a list context or narrow your search using the -types or
-class arguments

STACK Bio::DB::SeqFeature::Store::segment
/usr/local/share/perl5/Bio/DB/SeqFeature/Store.pm:1359
STACK toplevel bin/prepare-refseqs.pl:156
-------------------------------------------

Any idea how I can get around this? Thanks

On Fri, Feb 10, 2012 at 12:34 PM, Robert Buels <rb...@gm...> wrote:

> Hi Heewook,
>
> The best thing to do would be to write a little JSON conf file specifying
> how you want the tracks to be displayed, then use bin/biodb-to-json.pl to
> format them into your data/ directory.
>
> See http://jbrowse.org/code/**jbrowse-master/docs/tutorial/<http://jbrowse.org/code/jbrowse-master/docs/tutorial/>for details.
>
> Let me know if you have any more questions!
>
> Rob
>
>
> On 02/10/2012 11:54 AM, Heewook Lee wrote:
>
>> Hi, I got all the components installed and jbrowse is running. I am
>> trying to load ecoli onto jbrowse from NCBI files. I downloaded their
>> gff file and the reference sequence in fasta format. What's the best way
>> to get this set up?
>>
>

Re: [Gmod-ajax] setting up jbrowse with a bacterial reference genome

[Robert B. <rb...@gm...>]

Hi Heewook,

The best thing to do would be to write a little JSON conf file 
specifying how you want the tracks to be displayed, then use 
bin/biodb-to-json.pl to format them into your data/ directory.

See http://jbrowse.org/code/jbrowse-master/docs/tutorial/ for details.

Let me know if you have any more questions!

Rob

On 02/10/2012 11:54 AM, Heewook Lee wrote:
> Hi, I got all the components installed and jbrowse is running. I am
> trying to load ecoli onto jbrowse from NCBI files. I downloaded their
> gff file and the reference sequence in fasta format. What's the best way
> to get this set up?

Re: [Gmod-ajax] setting up jbrowse with a bacterial reference genome

[Robert B. <rb...@gm...>]

Hi Heewook,

The best thing to do would be to write a little JSON conf file 
specifying how you want the tracks to be displayed, then use 
bin/biodb-to-json.pl to format them into your data/ directory.

See http://jbrowse.org/code/jbrowse-master/docs/tutorial/ for details.

Let me know if you have any more questions!

Rob


On 02/10/2012 11:54 AM, Heewook Lee wrote:
> Hi, I got all the components installed and jbrowse is running. I am
> trying to load ecoli onto jbrowse from NCBI files. I downloaded their
> gff file and the reference sequence in fasta format. What's the best way
> to get this set up?
>
> -hl
>
>
>
> ------------------------------------------------------------------------------
> Virtualization&  Cloud Management Using Capacity Planning
> Cloud computing makes use of virtualization - but cloud computing
> also focuses on allowing computing to be delivered as a service.
> http://www.accelacomm.com/jaw/sfnl/114/51521223/
>
>
>
> _______________________________________________
> Gmod-ajax mailing list
> Gmo...@li...
> https://lists.sourceforge.net/lists/listinfo/gmod-ajax

[Gmod-ajax] setting up jbrowse with a bacterial reference genome

[Heewook L. <he...@um...>]

Hi, I got all the components installed and jbrowse is running. I am trying
to load ecoli onto jbrowse from NCBI files. I downloaded their gff file and
the reference sequence in fasta format. What's the best way to get this set
up?

-hl