BUG: using relative paths to indexed fasta file as genome fails to load correctly

127 views
Skip to first unread message

Malcolm Cook

unread,
Aug 23, 2016, 12:03:56 PM8/23/16
to igv-help
I believe there is a regression in support for this recent feature:

     "A file path or URL to an indexed fasta or .genome file can be used in place of the genome id.   The path can be absolute or relative to the parent directory of the session file."


I find in IGV 2.3.x, and last nights build, under both linux and windows, that creating a session file with all paths relative, files to load correctly with "File > Open Session" of "File > Load From URL"

It appears that the session files is erroneously treated as a .genome file!!!

For starters, the log reports "Loading genome..." on the path to the session file, and even tries to index it, creating a bogus empty .fai file and reporting an error trying to index it.

Further downstream errors ensue.  See the full log below.

This happens when the session uses relative pathname throughout.  I.e., the file begins with:

<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<Session genome="scaffolds.fa" hasGeneTrack="false" hasSequenceTrack="true" locus="ScGoXVd_5961:578700-600275" version="8">

and all the id=  of all the Tracks are similarly relative.

However, if I change the genome= to contain the full path to the fasta, all is good and it works.  I do not need to make the id= attributes full pathnames.... relative works here.

Alas, as a result, sessions are not portable.  I can no longer serve the session both over the web and over multiple mount points on different (window/mac/linux) file systems.

As always, thanks for IGV,

Malcolm


Log: of "File > Open Session"


INFO [2016-08-23 10:49:28,978]  [GenomeManager.java:145] [pool-4-thread-7]  Loading genome: S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.d2\aa_igv_session.xml
INFO [2016-08-23 10:49:30,866]  [FastaUtils.java:66] [pool-4-thread-7]  Creating index file at S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.d2\aa_igv_session.xml.fai
ERROR [2016-08-23 10:49:30,872]  [MessageUtils.java:64] [pool-4-thread-7]  Could not create index file: S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.d2\aa_igv_session.xml.fai
org.broad.igv.exceptions.DataLoadException: An error occurred while accessing:    S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.d2\aa_igv_session.xml<br>Fasta file has uneven line lengths in contig null
at org.broad.igv.feature.genome.FastaUtils.createIndexFile(FastaUtils.java:94)
at org.broad.igv.feature.genome.GenomeManager.loadFastaFile(GenomeManager.java:290)
at org.broad.igv.feature.genome.GenomeManager.loadGenome(GenomeManager.java:167)
at org.broad.igv.ui.IGV.loadGenome(IGV.java:699)
at org.broad.igv.ui.IGV.loadGenome(IGV.java:680)
at org.broad.igv.session.IGVSessionReader.processRootNode(IGVSessionReader.java:404)
at org.broad.igv.session.IGVSessionReader.loadSession(IGVSessionReader.java:324)
at org.broad.igv.ui.IGV.restoreSessionSynchronous(IGV.java:1484)
at org.broad.igv.ui.IGV$10.run(IGV.java:1453)
at org.broad.igv.util.LongRunningTask.call(LongRunningTask.java:72)
at java.util.concurrent.FutureTask.run(Unknown Source)
at java.util.concurrent.ThreadPoolExecutor.runWorker(Unknown Source)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(Unknown Source)
at java.lang.Thread.run(Unknown Source)
ERROR [2016-08-23 10:49:30,873]  [MessageUtils.java:74] [pool-4-thread-7]  Could not create index file: S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.d2\aa_igv_session.xml.fai
INFO [2016-08-23 10:49:35,952]  [GenomeManager.java:192] [pool-4-thread-7]  Genome loaded.  id= S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.d2\aa_igv_session.xml
INFO [2016-08-23 10:49:35,960]  [IGVCommandBar.java:307] [pool-4-thread-7]  Loading S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.d2\aa_igv_session.xml  aa_igv_session.xml   S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.d2\aa_igv_session.xml
INFO [2016-08-23 10:49:35,960]  [IGVCommandBar.java:307] [pool-4-thread-7]  Loading S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.d2\aa_igv_session.xml  aa_igv_session.xml   S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.d2\aa_igv_session.xml
INFO [2016-08-23 10:49:35,965]  [TrackLoader.java:120] [Thread-46]  Loading resource, path S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.d2\scaffolds_HiRise_input.bed.gz
INFO [2016-08-23 10:49:35,965]  [TrackLoader.java:120] [Thread-47]  Loading resource, path S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.d2\hoxCluster.bed.gz
INFO [2016-08-23 10:49:35,965]  [TrackLoader.java:120] [Thread-48]  Loading resource, path S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.d2\coding_genes.gff3.gz
INFO [2016-08-23 10:49:36,154]  [TrackLoader.java:120] [pool-4-thread-7]  Loading resource, path S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.d2\chicagoLibrary2.all.f1F3852.liftover.sorted.bam
INFO [2016-08-23 10:49:36,251]  [MessageUtils.java:74] [pool-4-thread-7]  <html>File: S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.d2\chicagoLibrary2.all.f1F3852.liftover.sorted.bam<br>does not contain any sequence names which match the current genome.<br><br>File: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;ScGoXVd_1, ScGoXVd_10, ScGoXVd_100, ScGoXVd_1000,  ...<br>Genome: 
INFO [2016-08-23 10:49:38,271]  [IGVSessionReader.java:1079] [pool-4-thread-7]  Warning.  No tracks were found with id: chicagoLibrary2.all.f1F3852.liftover.sorted.bam_coverage in session file
INFO [2016-08-23 10:49:38,271]  [IGVSessionReader.java:1079] [pool-4-thread-7]  Warning.  No tracks were found with id: chicagoLibrary2.all.f1F3852.liftover.sorted.bam_junctions in session file
INFO [2016-08-23 10:49:38,272]  [IGVSessionReader.java:1079] [pool-4-thread-7]  Warning.  No tracks were found with id: chicagoLibrary2.all.f1F3852.liftover.sorted.bam in session file
INFO [2016-08-23 10:49:38,275]  [MessageUtils.java:74] [pool-4-thread-7]  Cannot find feature or locus: ScGoXVd_5961:578700-600275

2016-08-23_10-40-45.png

Jim Robinson

unread,
Aug 23, 2016, 2:13:04 PM8/23/16
to igv-...@googlegroups.com

That's pretty weird.  I'll look at it. 

--

---
You received this message because you are subscribed to the Google Groups "igv-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/493b1dda-9d9f-4f29-849b-fef6259a59d1%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Jim Robinson

unread,
Aug 23, 2016, 5:25:57 PM8/23/16
to igv-...@googlegroups.com
Malcolm, I think this is fixed now. I'm not sure its a regression
though, I don't see any way it could ever have worked before. Anyway
try the "snapshot" build, if you confirm its fixed I'll do another release.

Jim



On 8/23/16 9:03 AM, Malcolm Cook wrote:

Malcolm Cook

unread,
Aug 23, 2016, 8:05:30 PM8/23/16
to igv-...@googlegroups.com
Jim!

Perfect.  

"File > Load From URL" works with sessions
"File > Open Session" works
Drag & Drop works from windows desktop works
Drag & Drop works from linux desktop work (used to give null pointer exception)

Hurrah for Jim!

Thanks,

Malcolm

Agreed - it was not a regression... there has always been murk in this water.  Very happy it is cleared up now.



--

--- You received this message because you are subscribed to a topic in the Google Groups "igv-help" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/igv-help/ZL8ghjsQq7c/unsubscribe.
To unsubscribe from this group and all its topics, send an email to igv-help+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/39e312a3-5a23-c521-6549-cec788982f79%40broadinstitute.org.

Malcolm Cook

unread,
Aug 25, 2016, 2:48:30 PM8/25/16
to igv-help
Jim

perhaps unrelated, but, now when I load such session files in a fresh launch of igv snapshot , I get immediately an UnsupportedFlavorException: Unicode String in the log, and after navigating with ctrl-F a few times on the track hoxCluster.bed.gz the system sloooowws down, memory usage in lower RH corner creeps up and up and up, and no new entries in the log.  

Any ideas?

Log follows:


INFO [2016-08-25 13:41:36,447]  [CommandListener.java:106] [Thread-23]  Listening on port 60151
ERROR [2016-08-25 13:41:46,591]  [DataPanelContainer.java:230] [AWT-EventQueue-2]  java.awt.datatransfer.UnsupportedFlavorException: Unicode String
INFO [2016-08-25 13:41:47,006]  [GenomeManager.java:145] [pool-4-thread-1]  Loading genome: S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.071516\Lamprey_final_assembly_07_15_2016.fa
INFO [2016-08-25 13:41:47,052]  [GenomeManager.java:192] [pool-4-thread-1]  Genome loaded.  id= S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.071516\Lamprey_final_assembly_07_15_2016.fa
INFO [2016-08-25 13:41:47,057]  [IGVCommandBar.java:307] [pool-4-thread-1]  Loading S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.071516\Lamprey_final_assembly_07_15_2016.fa  Lamprey_final_assembly_07_15_2016.fa   S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.071516\Lamprey_final_assembly_07_15_2016.fa
INFO [2016-08-25 13:41:47,058]  [IGVCommandBar.java:307] [pool-4-thread-1]  Loading S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.071516\Lamprey_final_assembly_07_15_2016.fa  Lamprey_final_assembly_07_15_2016.fa   S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.071516\Lamprey_final_assembly_07_15_2016.fa
INFO [2016-08-25 13:41:47,107]  [TrackLoader.java:120] [Thread-24]  Loading resource, path S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.071516\coding_genes.gff3.gz
INFO [2016-08-25 13:41:47,107]  [TrackLoader.java:120] [Thread-25]  Loading resource, path S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.071516\PMZ_v3_genes.gff3.gz
INFO [2016-08-25 13:41:47,107]  [TrackLoader.java:120] [Thread-26]  Loading resource, path S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.071516\liftover_PMZ_v1_genes.gff3.gz
INFO [2016-08-25 13:41:47,107]  [TrackLoader.java:120] [Thread-27]  Loading resource, path S:\Bioinformatics\analysis\lmw\hox_in_lamprey\pmg1.071516\hoxCluster.bed.gz

 

Jim Robinson

unread,
Aug 25, 2016, 3:04:43 PM8/25/16
to igv-...@googlegroups.com

You've probably got an invalid character (non-ascii) in one of your files, perhaps your session file.  That shouldn't cause a memory issue however.   I can't really help further without access to these files.

Are the files indexed with tabix?
--

---
You received this message because you are subscribed to the Google Groups "igv-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+u...@googlegroups.com.

Malcolm Cook

unread,
Aug 25, 2016, 3:48:01 PM8/25/16
to igv-...@googlegroups.com
yes - they are all indexed with tabix


Using last nights snapshot, try "File > Open URL" 

So far, so good.

Select the first track.

Type control-F a bunch of times (nine or ten).

IGV gets real sluggish and finally almost stops responding at around:

Inline image 1

I think this is perhaps related with another issue you improved recently at my report having to do with control-F and control-B paging to the next scaffold/chromosome...

Actually, the memory used is jumping around now between 125MB and 188MB but nothing is happening.  

Garbage collection run amok?  I'm used to that in emacs (doh), but not IGV.

Help?,

Malcolm


To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+unsubscribe@googlegroups.com.

--

---
You received this message because you are subscribed to a topic in the Google Groups "igv-help" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/igv-help/ZL8ghjsQq7c/unsubscribe.
To unsubscribe from this group and all its topics, send an email to igv-help+unsubscribe@googlegroups.com.

Jim Robinson

unread,
Aug 25, 2016, 6:09:03 PM8/25/16
to igv-...@googlegroups.com

I was able to provoke an error, which might be causing the slowness you experienced.   Try it again now (with updated snapshot build).

Malcolm Cook

unread,
Aug 25, 2016, 6:41:04 PM8/25/16
to igv-...@googlegroups.com
It appears somewhat better.

I can now Control-F more times before IGV memory indicator begins to jump up and down and IGV becomes extremely sluggish.  No errors or messages appear in console after initial session and track load.

Can I report more/better?

Thanks

Jim Robinson

unread,
Aug 25, 2016, 6:43:20 PM8/25/16
to igv-...@googlegroups.com

How does this behavior compare with using the released version?   The memory numbers you quote, BTW,  are all very small and shouldn't cause any issues.  How much memory are you allocating on startup (or, to rephrase,  which distribution package are you using?)

--

---
You received this message because you are subscribed to the Google Groups "igv-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/CAAAQmVYjBVgym6JtNBpRuVrbBD2v9XgjjePQOQD%2Bq%2BxQMF05sg%40mail.gmail.com.

Malcolm Cook

unread,
Aug 25, 2016, 6:48:57 PM8/25/16
to igv-...@googlegroups.com
This happens with both 750MB and 1.2 GB development snapshot versions.

The released version does not work with the data I have posted since the session uses relative paths, and the "genome" is just a plain fasta file (with corresponding .fai file in same directory).

Do you want me to 
 ? create a igv genome and try it with that?
 ? use fewer "chromosomes" in my genome instead of the 12 thousand scaffolds I've got?
 ?

~Malcolm


On Thu, Aug 25, 2016 at 5:43 PM, Jim Robinson <jrob...@broadinstitute.org> wrote:

How does this behavior compare with using the released version?   The memory numbers you quote, BTW,  are all very small and shouldn't cause any issues.  How much memory are you allocating on startup (or, to rephrase,  which distribution package are you using?)



On 8/25/16 3:40 PM, Malcolm Cook wrote:
It appears somewhat better.

I can now Control-F more times before IGV memory indicator begins to jump up and down and IGV becomes extremely sluggish.  No errors or messages appear in console after initial session and track load.

Can I report more/better?

Thanks
--

---
You received this message because you are subscribed to the Google Groups "igv-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+unsubscribe@googlegroups.com.

--

---
You received this message because you are subscribed to a topic in the Google Groups "igv-help" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/igv-help/ZL8ghjsQq7c/unsubscribe.
To unsubscribe from this group and all its topics, send an email to igv-help+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/d445c42b-b5cd-46f9-7a2e-93a74952648a%40broadinstitute.org.

Jim Robinson

unread,
Aug 25, 2016, 6:51:10 PM8/25/16
to igv-...@googlegroups.com

Those are both very small allocations, are you not running 64 bit windows yet?   You could try the packaged window app,  which bundles a 64 bit Java.  

RE released version, it should handle and absolute path to a fasta,  or does that not work either?

To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/CAAAQmVawtb92MsdppLrCSiBfDYDyenXnyP0YO2D8typHNYh-wQ%40mail.gmail.com.

Malcolm Cook

unread,
Aug 25, 2016, 8:23:46 PM8/25/16
to igv-...@googlegroups.com
Hi,

I am running Windows 7 64 bit and CentOS 6 64 bit

(I've got a Mac here somewhere too, but, hey).

I am having trouble ensuring that the 64-bit version of java is being used (under windows).  Thus I have not (yet) been able to use the large memory versions of IGV (igv_lm.jnlp) under web start in Windows.  When I try to run them, I get no message, Java launches, shows me

Inline image 1

and quits.

However, at your suggestion, I am able to run IGV under 2GB jvm using the packaged windows app.  However

 - it does NOT handle a relative path to a fasta (which I should expect since you just fixed this for me - I think)
 - when using an absolute path
   - it does not correctly skip between scaffolds while navigating between features using control-F
   - thus I don't know if the problem wrought by navigating forward 10 times is present, since I can not navigate forward at all using control-F

Turning to linux, running the snapshot with java 1.8.0_92 with 6GB I get the same sluggishness after control-F around 10 times. Memory starts at 161M of 1649M, loading the track using a relative path to a .genome file and memory jumps to 528M of 2,315M, hitting control-F and control-B a bunch of times, and IGV pretty much freezes up and memory starts skipping around between 545M and 2,149M or 2,96... (number is clipped).

??,

Malcolm




--

---
You received this message because you are subscribed to a topic in the Google Groups "igv-help" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/igv-help/ZL8ghjsQq7c/unsubscribe.
To unsubscribe from this group and all its topics, send an email to igv-help+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/91dd6168-a8ad-969d-dfb5-32d7656c36e1%40broadinstitute.org.

Malcolm Cook

unread,
Aug 25, 2016, 8:32:01 PM8/25/16
to igv-...@googlegroups.com
Jim,  I just notice that my linux session produced an error.  Perhaps a clue?


ERROR [2016-08-25 19:22:32,790]  [DefaultExceptionHandler.java:49] [AWT-EventQueue-0]  Unhandled exception
java.lang.ArrayIndexOutOfBoundsException: 37450
    at htsjdk.tribble.readers.TabixReader.reg2bins(TabixReader.java:153)
    at htsjdk.tribble.readers.TabixReader.query(TabixReader.java:417)
    at htsjdk.tribble.TabixFeatureReader.query(TabixFeatureReader.java:120)
    at org.broad.igv.feature.tribble.CachingFeatureReader.queryRaw(CachingFeatureReader.java:73)
    at org.broad.igv.feature.AbstractCacher.loadTiles(AbstractCacher.java:198)
    at org.broad.igv.feature.AbstractCacher.getBins(AbstractCacher.java:161)
    at org.broad.igv.feature.AbstractCacher.queryCached(AbstractCacher.java:102)
    at org.broad.igv.feature.tribble.CachingFeatureReader.query(CachingFeatureReader.java:104)
    at org.broad.igv.track.TribbleFeatureSource$IndexedFeatureSource.getFeatures(TribbleFeatureSource.java:192)
    at org.broad.igv.track.FeatureTrackUtils.nextFeature(FeatureTrackUtils.java:75)
    at org.broad.igv.track.FeatureTrack.nextFeature(FeatureTrack.java:1004)
    at org.broad.igv.ui.event.GlobalKeyDispatcher.nextFeature(GlobalKeyDispatcher.java:401)
    at org.broad.igv.ui.event.GlobalKeyDispatcher.access$000(GlobalKeyDispatcher.java:61)
    at org.broad.igv.ui.event.GlobalKeyDispatcher$1.actionPerformed(GlobalKeyDispatcher.java:116)
    at org.broad.igv.ui.event.GlobalKeyDispatcher$EnableWrappedAction.actionPerformed(GlobalKeyDispatcher.java:446)
    at org.broad.igv.ui.event.GlobalKeyDispatcher.dispatchKeyEvent(GlobalKeyDispatcher.java:94)
    at java.awt.DefaultKeyboardFocusManager.preDispatchKeyEvent(DefaultKeyboardFocusManager.java:1068)
    at java.awt.DefaultKeyboardFocusManager.typeAheadAssertions(DefaultKeyboardFocusManager.java:945)
    at java.awt.DefaultKeyboardFocusManager.dispatchEvent(DefaultKeyboardFocusManager.java:771)
    at java.awt.Component.dispatchEventImpl(Component.java:4760)
    at java.awt.Container.dispatchEventImpl(Container.java:2294)
    at java.awt.Window.dispatchEventImpl(Window.java:2746)
    at java.awt.Component.dispatchEvent(Component.java:4711)
    at java.awt.EventQueue.dispatchEventImpl(EventQueue.java:758)
    at java.awt.EventQueue.access$500(EventQueue.java:97)
    at java.awt.EventQueue$3.run(EventQueue.java:709)
    at java.awt.EventQueue$3.run(EventQueue.java:703)
    at java.security.AccessController.doPrivileged(Native Method)
    at java.security.ProtectionDomain$JavaSecurityAccessImpl.doIntersectionPrivilege(ProtectionDomain.java:76)
    at java.security.ProtectionDomain$JavaSecurityAccessImpl.doIntersectionPrivilege(ProtectionDomain.java:86)
    at java.awt.EventQueue$4.run(EventQueue.java:731)
    at java.awt.EventQueue$4.run(EventQueue.java:729)
    at java.security.AccessController.doPrivileged(Native Method)
    at java.security.ProtectionDomain$JavaSecurityAccessImpl.doIntersectionPrivilege(ProtectionDomain.java:76)
    at java.awt.EventQueue.dispatchEvent(EventQueue.java:728)
    at java.awt.EventDispatchThread.pumpOneEventForFilters(EventDispatchThread.java:201)
    at java.awt.EventDispatchThread.pumpEventsForFilter(EventDispatchThread.java:116)
    at java.awt.EventDispatchThread.pumpEventsForHierarchy(EventDispatchThread.java:105)
    at java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:101)
    at java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:93)
    at java.awt.EventDispatchThread.run(EventDispatchThread.java:82)




Jim Robinson

unread,
Aug 26, 2016, 12:04:35 AM8/26/16
to igv-...@googlegroups.com

Yes, that is a valuable clue.   The magic bin number "37450"  was causing problems in igv.js,  I discovered it is not a real index bin but a place to shove meta data. Could you share that tabix file and its index (or is it one of the files in the session you shared already)?

Jim

To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/CAAAQmVbqVs%2B-mMSbrpWXAw7-f%3DSq1vWMko-4KuJxBVizxhTqEw%40mail.gmail.com.

Jim Robinson

unread,
Aug 26, 2016, 12:41:33 AM8/26/16
to igv-...@googlegroups.com

Actually looking deeper this is strange.  37450 is beyond the range of allowed bins in a tabix indexed file.     Its a little hard to see how you get this unless your contig length is > 2^29-1.   Is that possible?  If so it can't be indexed with tabix.

I'll need access to this file,  then I can figure out what to report to the htsjdk team.

To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/CAAAQmVbqVs%2B-mMSbrpWXAw7-f%3DSq1vWMko-4KuJxBVizxhTqEw%40mail.gmail.com.

Malcolm Cook

unread,
Aug 26, 2016, 9:32:43 AM8/26/16
to igv-...@googlegroups.com
Jim,

The files are all here: http://research.stowers.org/mec/PMZ_v3_Review/ (including the session, the .gz files, the .gz.tbi files, and the uncompressed gff files which you probably don't need).

However, and FYI, I am home trying this on my 64-bit mac with 64-bit up-to-date java running IGV Snapshot Large Memory and scanning through the 2nd track (coding_genes liftover from Lamprey2) by holding down control-F which causes memory use to jump to 731MB of 835MB, the cursor to spin, and the app to become non-responsive, but I did not get any associated error in the Java console.  After 10 minutes, I have to use "Apple > Force Quit".

~Malcolm

--

---
You received this message because you are subscribed to a topic in the Google Groups "igv-help" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/igv-help/ZL8ghjsQq7c/unsubscribe.
To unsubscribe from this group and all its topics, send an email to igv-help+unsubscribe@googlegroups.com.

Malcolm Cook

unread,
Aug 26, 2016, 9:36:49 AM8/26/16
to igv-...@googlegroups.com
my largest scaffold is

scaf_00001	27620446

much smaller than 2**29 


--

---
You received this message because you are subscribed to a topic in the Google Groups "igv-help" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/igv-help/ZL8ghjsQq7c/unsubscribe.
To unsubscribe from this group and all its topics, send an email to igv-help+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/e221956e-6634-867e-c2b1-c976728450b7%40broadinstitute.org.

Jim Robinson

unread,
Aug 26, 2016, 9:56:55 AM8/26/16
to igv-...@googlegroups.com
Holding down the ctrl-F key could overwhelm memory because you are continually firing off http requests to load data before the previous request has completed.  Are you at least waiting for the screen to repaint?   I might look into making the key inoperable until the previous load has completed.
To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/CAAAQmVaAmWxODeVyVuC1JH1cUO59J-%3D_i3UV04-LT3t_B5O1ow%40mail.gmail.com.

Cook, Malcolm

unread,
Aug 26, 2016, 9:58:14 AM8/26/16
to igv-...@googlegroups.com

This happens even if I wait for the screen to repaint.  I just happens faster if I hold the key down.

Jim,

 

Hi,

 

nline image 1

and quits.

 

To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+u...@googlegroups.com.

--

---
You received this message because you are subscribed to a topic in the Google Groups "igv-help" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/igv-help/ZL8ghjsQq7c/unsubscribe.

To unsubscribe from this group and all its topics, send an email to igv-help+u...@googlegroups.com.


For more options, visit https://groups.google.com/d/optout.

--

---
You received this message because you are subscribed to the Google Groups "igv-help" group.

To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+u...@googlegroups.com.


For more options, visit https://groups.google.com/d/optout.

--

---
You received this message because you are subscribed to a topic in the Google Groups "igv-help" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/igv-help/ZL8ghjsQq7c/unsubscribe.

To unsubscribe from this group and all its topics, send an email to igv-help+u...@googlegroups.com.


For more options, visit https://groups.google.com/d/optout.

--

---
You received this message because you are subscribed to the Google Groups "igv-help" group.

To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+u...@googlegroups.com.


For more options, visit https://groups.google.com/d/optout.

--

---
You received this message because you are subscribed to a topic in the Google Groups "igv-help" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/igv-help/ZL8ghjsQq7c/unsubscribe.

To unsubscribe from this group and all its topics, send an email to igv-help+u...@googlegroups.com.


For more options, visit https://groups.google.com/d/optout.

--

---
You received this message because you are subscribed to the Google Groups "igv-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/CAAAQmVaAmWxODeVyVuC1JH1cUO59J-%3D_i3UV04-LT3t_B5O1ow%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

 

--


---
You received this message because you are subscribed to a topic in the Google Groups "igv-help" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/igv-help/ZL8ghjsQq7c/unsubscribe.

To unsubscribe from this group and all its topics, send an email to igv-help+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/6ecfc50a-ba00-64ed-57b8-d8b3d25c868e%40broadinstitute.org.

Jim Robinson

unread,
Aug 26, 2016, 11:50:29 PM8/26/16
to igv-...@googlegroups.com
Malcolm,

Can you try the snapshot build again? I made a change that improves it
greatly for me, but could you confirm? If you find it fixed it will
confirm I am on the right path, conversely I will look elsewhere.

Thanks

Jim

Cook, Malcolm

unread,
Aug 27, 2016, 12:23:19 AM8/27/16
to igv-...@googlegroups.com
It is much much improved!

Is still managed to get a “Error loading features for interval: scaf_00005:8116714-8540335 java.util.ConcurrentModificationException” but this was after sitting on ctl-F for a few large scaffolds. And IGV recovered without freeze.

But, the behavior of contro-f now appears different. Instead of nav to next feature, it ppears to go forward to next currently not displayed feature. Is this by (re)design?

I sure can live with it, if so….
~m
--

---
You received this message because you are subscribed to a topic in the Google Groups "igv-help" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/igv-help/ZL8ghjsQq7c/unsubscribe.
To unsubscribe from this group and all its topics, send an email to igv-help+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/f03052cb-0d80-1985-6bf6-dd083855d71e%40broadinstitute.org.

Jim Robinson

unread,
Aug 27, 2016, 1:06:32 AM8/27/16
to igv-...@googlegroups.com
RE ctrl-f change, that's not by design, and are your sure you didn't
have a few ctrl-Fs "in the queue" that were just catching up.

I would be interested in seeing where that concurent modifcation
exception occured, it should be in the log.

Malcolm Cook

unread,
Aug 27, 2016, 12:17:09 PM8/27/16
to igv-...@googlegroups.com
oops, ctrl-f does navigate as before - pardon my mistake - and quite quickly and robustly

i am now unable to cause the concurrent modification exception....

all testing done on mac - will test Monday on other platforms... otherwise, I think this closes this matter.... 

as usual, thanks for the responsive consideration of my issues

On Sat, Aug 27, 2016 at 12:06 AM, Jim Robinson <jrob...@broadinstitute.org> wrote:
RE ctrl-f change,  that's not by design, and are your sure you didn't have a few ctrl-Fs "in the queue" that were just catching up.

I would be interested in seeing where that concurent modifcation exception occured, it should be in the log.



On 8/26/16 9:23 PM, Cook, Malcolm wrote:
It is much much improved!

Is still managed to get a “Error loading features for interval: scaf_00005:8116714-8540335 java.util.ConcurrentModificationException” but this was after sitting on ctl-F for a few large scaffolds.  And IGV recovered without freeze.

But, the behavior of contro-f now appears different.  Instead of nav to next feature, it ppears to go forward to next currently not displayed feature.  Is this by (re)design?

I sure can live with it, if so….
~m


On 8/26/16, 10:50 PM, "Jim Robinson" <igv-...@googlegroups.com on behalf of jrob...@broadinstitute.org> wrote:

     Malcolm,
          Can you try the snapshot build again?  I made a change that improves it
     greatly for me, but could you confirm?   If you find it fixed it will
     confirm I am on the right path, conversely I will look elsewhere.
          Thanks
          Jim
          --
          ---
     You received this message because you are subscribed to a topic in the Google Groups "igv-help" group.
     To unsubscribe from this topic, visit https://groups.google.com/d/topic/igv-help/ZL8ghjsQq7c/unsubscribe.
     To unsubscribe from this group and all its topics, send an email to igv-help+unsubscribe@googlegroups.com.

     To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/f03052cb-0d80-1985-6bf6-dd083855d71e%40broadinstitute.org.
     For more options, visit https://groups.google.com/d/optout.
     


--

--- You received this message because you are subscribed to a topic in the Google Groups "igv-help" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/igv-help/ZL8ghjsQq7c/unsubscribe.
To unsubscribe from this group and all its topics, send an email to igv-help+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/58761d5c-eade-8b17-f791-eb7b0276657c%40broadinstitute.org.

Jim Robinson

unread,
Aug 27, 2016, 6:09:12 PM8/27/16
to igv-...@googlegroups.com

OK.  If you see the concurrent modification exception again please open another ticket,  and include the stack trace from the log.  

Jim

Reply all
Reply to author
Forward
0 new messages