Tophat: segment-based junction search failed with err =-11

jdanderson · 07-19-2011, 07:02 AM

Hello Everyone,

I was wondering if anyone has ever gotten or knows the source of the following error. I'm running Tophat (v1.3.1) on a dual core, 16GB RAM (older) desktop. The following was my command and my output while running a new paired end dataset.

Quote:

tophat -o /home/johnathon/topout/ -r 209 --mate-std-dev 89 --solexa1.3-quals -p 2 -G /home/johnathon/ucsc_usable_ens_mus_ncbi37_63.gtf /home/johnathon/bowtie-0.12.7/indexes/mm9 /home/johnathon/raw_seq_data/DP_PEsequence1.txt /home/johnathon/raw_seq_data/DP_PEsequence2.txt

[Mon Jul 18 10:46:00 2011] Beginning TopHat run (v1.3.1)
-----------------------------------------------
[Mon Jul 18 10:46:00 2011] Preparing output location /home/johnathon/topout//
[Mon Jul 18 10:46:00 2011] Checking for Bowtie index files
[Mon Jul 18 10:46:00 2011] Checking for reference FASTA file
[Mon Jul 18 10:46:00 2011] Checking for Bowtie
Bowtie version: 0.12.7.0
[Mon Jul 18 10:46:00 2011] Checking for Samtools
Samtools Version: 0.1.17
[Mon Jul 18 10:46:00 2011] Generating SAM header for /home/johnathon/bowtie-0.12.7/indexes/mm9
[Mon Jul 18 10:46:16 2011] Preparing reads
format: fastq
quality scale: phred64 (reads generated with GA pipeline version >= 1.3)
[Mon Jul 18 10:46:16 2011] Reading known junctions from GTF file
Left reads: min. length=107, count=28882563
Right reads: min. length=107, count=28898589
[Mon Jul 18 11:13:30 2011] Mapping left_kept_reads against mm9 with Bowtie

[Mon Jul 18 12:29:44 2011] Processing bowtie hits
[Mon Jul 18 12:52:49 2011] Mapping left_kept_reads_seg1 against mm9 with Bowtie (1/4)
[Mon Jul 18 13:53:14 2011] Mapping left_kept_reads_seg2 against mm9 with Bowtie (2/4)
[Mon Jul 18 14:55:12 2011] Mapping left_kept_reads_seg3 against mm9 with Bowtie (3/4)
[Mon Jul 18 15:56:44 2011] Mapping left_kept_reads_seg4 against mm9 with Bowtie (4/4)
[Mon Jul 18 16:20:58 2011] Mapping right_kept_reads against mm9 with Bowtie
[Mon Jul 18 17:36:52 2011] Processing bowtie hits
[Mon Jul 18 18:00:22 2011] Mapping right_kept_reads_seg1 against mm9 with Bowtie (1/4)
[Mon Jul 18 19:02:18 2011] Mapping right_kept_reads_seg2 against mm9 with Bowtie (2/4)
[Mon Jul 18 20:04:43 2011] Mapping right_kept_reads_seg3 against mm9 with Bowtie (3/4)
[Mon Jul 18 21:07:18 2011] Mapping right_kept_reads_seg4 against mm9 with Bowtie (4/4)
[Mon Jul 18 21:31:13 2011] Searching for junctions via segment mapping
[FAILED]
Error: segment-based junction search failed with err =-11

The following is the output of my segment_juncs.log file from the run:

Quote:

segment_juncs v1.3.1 (2443)
---------------------------
Loading reference sequences...
Loading chr1...done
Loading left segment hits...
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record
Error in accessing sequence record

Any ideas on what might have caused this?

Cheers,
Johnathon

Suimye · 05-17-2012, 09:42 PM

I also got this message when Im running TopHat v2.0.0 (Bowtie version:2.0.0.5).
Do you have any solution for this error?

Suimye

jdanderson · 05-17-2012, 09:51 PM

Sorry, I actually can't recall how I resolved this error, although I wasn't using Tophat v2.0 when I ran into it.

Maybe Cole, or another Tophat programmer could be of more help. It also might help if you posted more info about the run you were attempting to perform and what preceeded the error message.

Cheers,
Johnathon

Suimye · 05-17-2012, 10:54 PM

Hi, Johnathon

Thank you for your rapidly replying and suggestion.
I am just trying by using the older version of Tophat and bowtie.

Sincerely,
Suimye

sameet · 06-11-2012, 12:31 PM

I am using the latest tophat version, but the error persists. I am also trying with the oldest version available in our system to see if that works.

sameet · 06-11-2012, 01:55 PM

Quote:

[Mon Jun 11 15:35:24 2012] Beginning TopHat run (v1.4.0)
-----------------------------------------------
[Mon Jun 11 15:35:24 2012] Preparing output location wt_Deli_thout/
[Mon Jun 11 15:35:24 2012] Checking for Bowtie index files
[Mon Jun 11 15:35:24 2012] Checking for reference FASTA file
[Mon Jun 11 15:35:24 2012] Checking for Bowtie
Bowtie version: 0.12.5.0
[Mon Jun 11 15:35:24 2012] Checking for Samtools
Samtools Version: 0.1.18
[Mon Jun 11 15:35:24 2012] Generating SAM header for /fdb/bowtie/indexes/genome
format: fastq
quality scale: phred33 (default)
[Mon Jun 11 15:35:24 2012] Preparing reads
left reads: min. length=45, count=9213949
right reads: min. length=42, count=7830358
[Mon Jun 11 15:41:45 2012] Mapping left_kept_reads against genome with Bowtie
[Mon Jun 11 16:06:18 2012] Processing bowtie hits
[Mon Jun 11 16:12:37 2012] Mapping left_kept_reads_seg1 against genome with Bowtie (1/2)
[Mon Jun 11 16:13:57 2012] Mapping left_kept_reads_seg2 against genome with Bowtie (2/2)
[Mon Jun 11 16:16:01 2012] Mapping right_kept_reads against genome with Bowtie
[Mon Jun 11 16:39:09 2012] Processing bowtie hits
[Mon Jun 11 16:44:42 2012] Mapping right_kept_reads_seg1 against genome with Bowtie (1/1)
[Mon Jun 11 16:53:43 2012] Searching for junctions via segment mapping
[FAILED]
Error: segment-based junction search failed with err =-11
Loading left segment hits...

I still get the same error with even older release of tophat.

dchen · 06-12-2012, 06:02 AM

I received the same error message, with Tophat 2.0.0 + Bowtie2.0.0.6,
but could fix it by correcting the Bowtie genome index, in my case hg19.
While the downloaded index contained the random chromosomes, the hg19.fa fasta file did not. After re-building the index without the random chromosomes,
the error message disappeared.

Hoping that this was helpful,
Doris

sameet · 06-12-2012, 06:20 AM

Thanks Doris, I will try to rebuild the index and see if that helps. I will update the post accordingly.

hmortens · 08-23-2012, 09:03 AM

How did you re-build the index without the random chromosomes????

bruce01 · 01-09-2013, 09:59 AM

By random chromosomes do you mean scaffolds (eg chrG*)? Do you just remove all from ref.fasta and build your index from this? I thought I had them removed.

Does anyone know specifically what is causing the error?

apredeus · 08-08-2013, 08:53 AM

I also get the error with the same ID, with a very established tophat 1.4.1.1

These are shorter reads I'm working with (25 bp), so I wonder it that is the problem?

I dug in the code a bit but couldn't find what exactly "-11" refers to. I wish there'd be a more verbose mode to tophat that would give more information on errors.

Anyhow, if anybody knows more about this, I'd love to know it.

PS yeah, and segment_juncs.log has no error at all (and it's the last log written). It looks like the program just stopped.

apredeus · 08-08-2013, 09:14 AM

and the command I use is the following:

tophat -p 8 -G $DATA/mm9.mrna.10.31.gtf --prefilter-multihits -o $WDIR/tophat_$TAG --segment-length 12 --max-multihits 15 $BOWTIE/mm9 $FQFILE

alabadorf · 02-04-2014, 09:33 AM

I encountered this error too and it was indeed related to a mismatch between the index and fasta file. Instead of rebuilding the index without random chromosomes, I used bowtie2-inspect to create a new fasta file from the index. The error went away.

The exit status code -11 corresponds to a segmentation fault suffered by the segment_juncs program, where I suspect it tries to find a fasta sequence but can't. The source of the error is actually in the SeqAn distribution, which they have bundled with the tophat release.

1520191 · 04-23-2014, 04:49 AM

Quote:

Originally Posted by alabadorf

I encountered this error too and it was indeed related to a mismatch between the index and fasta file. Instead of rebuilding the index without random chromosomes, I used bowtie2-inspect to create a new fasta file from the index. The error went away.

The exit status code -11 corresponds to a segmentation fault suffered by the segment_juncs program, where I suspect it tries to find a fasta sequence but can't. The source of the error is actually in the SeqAn distribution, which they have bundled with the tophat release.

Hi,I would like to know how you use bowtie2-inspect create a new fasta file from the index?whats your command?
my reference file is :musref
my index file are : musref.1.bt2 musref.2.bt2 musref.3.bt2 musref.4.bt2 musref.rev.1.bt2 musref.rev.2.bt2
what command can help me create a .fa file?
Thank you a lot!

mastal · 04-23-2014, 05:01 AM

see the Bowtie 2 Manual,

the section titled "The bowtie2-inspect index inspector"

http://bowtie-bio.sourceforge.net/bo...ndex-inspector

here is a quote from the manual:

"bowtie2-inspect extracts information from a Bowtie index about what kind of index it is and what reference sequences were used to build it. When run without any options, the tool will output a FASTA file containing the sequences of the original references (with all non-A/C/G/T characters converted to Ns). It can also be used to extract just the reference sequence names using the -n/--names option or a more verbose summary using the -s/--summary option."

pepperoni · 10-14-2017, 07:47 AM

Hello none of the recommendations had worked for me, has anyone found out what is the problem with error -11 and how to solve it?
thanks

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Tophat error -segment-based junction search failled with err=1	upadhyayanm	Bioinformatics	24	06-21-2016 07:20 PM
segment-based junction search failed	Wei-HD	Bioinformatics	6	07-18-2013 03:58 AM
Tophat Error: Error: segment-based junction search failed with err =-6	sjnewhouse	RNA Sequencing	8	03-19-2013 05:14 AM
yet another "Error: segment-based junction search failed with err = -9"	liux	Bioinformatics	1	08-24-2010 10:48 AM
Error: segment-based junction search failed with err = -9	Albert Cheng	Bioinformatics	1	08-19-2010 09:24 AM

06-11-2012, 12:31 PM	#5
sameet Member Location: Earth Join Date: Apr 2010 Posts: 34	The error persists! I am using the latest tophat version, but the error persists. I am also trying with the oldest version available in our system to see if that works. __________________ Sameet Mehta (Ph.D.), Visiting Fellow, National Cancer Insitute, Bethesda, US.

06-12-2012, 06:20 AM	#8
sameet Member Location: Earth Join Date: Apr 2010 Posts: 34	Thanks Doris, I will try to rebuild the index and see if that helps. I will update the post accordingly. __________________ Sameet Mehta (Ph.D.), Visiting Fellow, National Cancer Insitute, Bethesda, US.

05-17-2012, 09:42 PM	#2
Suimye Junior Member Location: Tokyo Join Date: May 2010 Posts: 9	I also got this message when Im running TopHat v2.0.0 (Bowtie version:2.0.0.5). Do you have any solution for this error? Suimye

05-17-2012, 09:51 PM	#3
jdanderson Member Location: Davis, CA Join Date: Sep 2010 Posts: 45	Sorry, I actually can't recall how I resolved this error, although I wasn't using Tophat v2.0 when I ran into it. Maybe Cole, or another Tophat programmer could be of more help. It also might help if you posted more info about the run you were attempting to perform and what preceeded the error message. Cheers, Johnathon

05-17-2012, 10:54 PM	#4
Suimye Junior Member Location: Tokyo Join Date: May 2010 Posts: 9	Hi, Johnathon Thank you for your rapidly replying and suggestion. I am just trying by using the older version of Tophat and bowtie. Sincerely, Suimye

06-12-2012, 06:02 AM	#7
dchen Junior Member Location: Austria Join Date: Apr 2012 Posts: 1	I received the same error message, with Tophat 2.0.0 + Bowtie2.0.0.6, but could fix it by correcting the Bowtie genome index, in my case hg19. While the downloaded index contained the random chromosomes, the hg19.fa fasta file did not. After re-building the index without the random chromosomes, the error message disappeared. Hoping that this was helpful, Doris

08-23-2012, 09:03 AM	#9
hmortens Member Location: RTP Join Date: Aug 2011 Posts: 12	How did you re-build the index without the random chromosomes????

01-09-2013, 09:59 AM	#10
bruce01 Senior Member Location: . Join Date: Mar 2011 Posts: 157	By random chromosomes do you mean scaffolds (eg chrG*)? Do you just remove all from ref.fasta and build your index from this? I thought I had them removed. Does anyone know specifically what is causing the error?

08-08-2013, 08:53 AM	#11
apredeus Senior Member Location: Bioinformatics Institute, SPb Join Date: Jul 2012 Posts: 151	I also get the error with the same ID, with a very established tophat 1.4.1.1 These are shorter reads I'm working with (25 bp), so I wonder it that is the problem? I dug in the code a bit but couldn't find what exactly "-11" refers to. I wish there'd be a more verbose mode to tophat that would give more information on errors. Anyhow, if anybody knows more about this, I'd love to know it. PS yeah, and segment_juncs.log has no error at all (and it's the last log written). It looks like the program just stopped.

08-08-2013, 09:14 AM	#12
apredeus Senior Member Location: Bioinformatics Institute, SPb Join Date: Jul 2012 Posts: 151	and the command I use is the following: tophat -p 8 -G $DATA/mm9.mrna.10.31.gtf --prefilter-multihits -o $WDIR/tophat_$TAG --segment-length 12 --max-multihits 15 $BOWTIE/mm9 $FQFILE

02-04-2014, 09:33 AM	#13
alabadorf Junior Member Location: Boston Join Date: Jun 2010 Posts: 3	I encountered this error too and it was indeed related to a mismatch between the index and fasta file. Instead of rebuilding the index without random chromosomes, I used bowtie2-inspect to create a new fasta file from the index. The error went away. The exit status code -11 corresponds to a segmentation fault suffered by the segment_juncs program, where I suspect it tries to find a fasta sequence but can't. The source of the error is actually in the SeqAn distribution, which they have bundled with the tophat release.

04-23-2014, 05:01 AM	#15
mastal Senior Member Location: uk Join Date: Mar 2009 Posts: 667	see the Bowtie 2 Manual, the section titled "The bowtie2-inspect index inspector" http://bowtie-bio.sourceforge.net/bo...ndex-inspector here is a quote from the manual: "bowtie2-inspect extracts information from a Bowtie index about what kind of index it is and what reference sequences were used to build it. When run without any options, the tool will output a FASTA file containing the sequences of the original references (with all non-A/C/G/T characters converted to Ns). It can also be used to extract just the reference sequence names using the -n/--names option or a more verbose summary using the -s/--summary option."

10-14-2017, 07:47 AM	#16
pepperoni Member Location: US Join Date: Oct 2011 Posts: 59	Hello none of the recommendations had worked for me, has anyone found out what is the problem with error -11 and how to solve it? thanks