Tophat Error: Report generation failed with err = 1

[Darwin]

Hi,
I am new to the bioinformatics field but want to map my RNA-Seq data using TopHat. I've installed bowtie and tophat and the sample test on bowtie ran successfully (old powermac G5, 2GB RAM).

However, when I test Tophat with the test_data set I get the error message: Error: Report generation failed with err = 1.

Without knowing much about programing it seems to me that the -k parameter could be wrong. I suspect it should be "-k 1" whereas Tophat has it as "-k 1,1n". Anyone encountered this or have any ideas?

This is the output I get after the command line:

-----

Macintosh:~/test_data darwin$ tophat -r 20 test_ref reads_1.fq reads_2.fq

[Thu Feb 18 20:02:21 2010] Beginning TopHat run (v1.0.13)
-----------------------------------------------
[Thu Feb 18 20:02:21 2010] Preparing output location ./tophat_out/
[Thu Feb 18 20:02:21 2010] Checking for Bowtie index files
[Thu Feb 18 20:02:21 2010] Checking for reference FASTA file
[Thu Feb 18 20:02:21 2010] Checking for Bowtie
Bowtie version: 0.12.3.0
[Thu Feb 18 20:02:21 2010] Checking reads
seed length: 75bp
format: fastq
quality scale: phred33 (default)
[Thu Feb 18 20:02:21 2010] Mapping reads against test_ref with Bowtie
[Thu Feb 18 20:02:32 2010] Joining segment hits
Splitting reads into 3 segments
[Thu Feb 18 20:02:32 2010] Mapping reads against test_ref with Bowtie
[Thu Feb 18 20:02:44 2010] Mapping reads against test_ref with Bowtie
[Thu Feb 18 20:02:55 2010] Mapping reads against test_ref with Bowtie
[Thu Feb 18 20:03:06 2010] Mapping reads against test_ref with Bowtie
[Thu Feb 18 20:03:17 2010] Joining segment hits
Splitting reads into 3 segments
[Thu Feb 18 20:03:17 2010] Mapping reads against test_ref with Bowtie
[Thu Feb 18 20:03:29 2010] Mapping reads against test_ref with Bowtie
[Thu Feb 18 20:03:40 2010] Mapping reads against test_ref with Bowtie
[Thu Feb 18 20:03:51 2010] Searching for junctions via segment mapping
[Thu Feb 18 20:03:51 2010] Retrieving sequences for splices
[Thu Feb 18 20:03:51 2010] Indexing splices
[Thu Feb 18 20:03:53 2010] Mapping reads against segment_juncs with Bowtie
[Thu Feb 18 20:03:53 2010] Mapping reads against segment_juncs with Bowtie
[Thu Feb 18 20:03:53 2010] Mapping reads against segment_juncs with Bowtie
[Thu Feb 18 20:03:53 2010] Joining segment hits
sort: invalid field specification `-k 1,1n'
[Thu Feb 18 20:03:53 2010] Mapping reads against segment_juncs with Bowtie
[Thu Feb 18 20:03:53 2010] Mapping reads against segment_juncs with Bowtie
[Thu Feb 18 20:03:54 2010] Mapping reads against segment_juncs with Bowtie
[Thu Feb 18 20:03:54 2010] Joining segment hits
sort: invalid field specification `-k 1,1n'
[Thu Feb 18 20:03:54 2010] Reporting output tracks
[FAILED]
Error: Report generation failed with err = 1

------------

Cheers,
Darwin

[owomo]

I meet the same err message, can anybody help us?

tophat -p 10 -o ORYwthDEF -r -70 -G input.gff

[MQ-BCBB]

Hello,
Did you ever found the cause of the error?
Thank you,
Mariam.

[Uwe Appelt]

I also receive this error message with TopHat 1.0.14. I, however, get it for the first time and my read-config is somewhat unusual, because i have "very few" reads that poorly align to the reference genome. Console output looks like this:

~$ tophat --num-threads 8 --solexa1.3-quals Provirus reads.fq

[Wed Sep 15 22:37:15 2010] Beginning TopHat run (v1.0.14)
-----------------------------------------------
[Wed Sep 15 22:37:15 2010] Preparing output location ./tophat_out/
[Wed Sep 15 22:37:15 2010] Checking for Bowtie index files
[Wed Sep 15 22:37:15 2010] Checking for reference FASTA file
[Wed Sep 15 22:37:15 2010] Checking for Bowtie
    Bowtie version:         0.12.5.0
[Wed Sep 15 22:37:15 2010] Checking reads
    seed length:     50bp
    format:         fastq
    quality scale:     phred64 (reads generated with GA pipeline version >= 1.3)
[Wed Sep 15 22:37:56 2010] Mapping reads against Provirus with Bowtie
[Wed Sep 15 22:38:01 2010] Joining segment hits
[Wed Sep 15 22:38:01 2010] Searching for junctions via segment mapping
Warning: junction database is empty!
[Wed Sep 15 22:38:28 2010] Joining segment hits
[Wed Sep 15 22:38:28 2010] Reporting output tracks
[B]    [FAILED]
Error: Report generation failed with err = 1[/B]

I thought it's probably related to (underlying Bowtie) not finding any mappable reads, but thats not the case, because:

~$ bowtie -t --threads 8 --solexa1.3-quals Provirus reads.fq reads.bout.txt

Time loading forward index: 00:00:00
Time loading mirror index: 00:00:00
Seeded quality full-index search: 00:00:02
# reads processed: 651981
[B]# reads with at least one reported alignment: 315 (0.05%)[/B]
# reads that failed to align: 651666 (99.95%)
Reported 315 alignments to 1 output stream(s)
Time searching: 00:00:02
Overall time: 00:00:02

Any suggestions appreciated! Thanks in advance & Cheers
Uwe

[grahammcv]

I get the same error message as Uwe, but have been unable to find the cause, despite trying many different things. I have tried TopHat versions 1.0.14, 1.0.13, 1.0.12, and 1.0.11 and Bowtie versions 0.12.7.0, 0.12.5.0 and 0.12.3.0. I compiled TopHat and Bowtie myself but also tried the precompiled 64-bit linux binaries of the most recent versions. I ran TopHat on human 37 bp paired-end reads (with a mean separation argument [-r] of 100, 200, 300 bp) and on human 46 bp single reads. I tried with or without the butterfly-search argument. I set the segment-length argument 10 or 20. I modified the code so that the seed length is 10, 20 or 25 rather than choosing the minimum read length, which is what the TopHat code does by default. I verified that the quality scores are in phred33 format, which is also what TopHat chooses by default. I built my own hg18 bowtie indexes using the make_hg18.sh script and used the prebuilt indexes from the website. In all cases I get the same result: no junctions appear to be found. A typical result looks like:

tophat $BOWTIE_INDEXES/hg18 data/test_nb/NA19119_argonne.fastq 

[Fri Sep 17 09:13:00 2010] Beginning TopHat run (v1.0.14)
-----------------------------------------------
[Fri Sep 17 09:13:00 2010] Preparing output location ./tophat_out/
[Fri Sep 17 09:13:00 2010] Checking for Bowtie index files
[Fri Sep 17 09:13:00 2010] Checking for reference FASTA file
[Fri Sep 17 09:13:00 2010] Checking for Bowtie
    Bowtie version:         0.12.7.0
[Fri Sep 17 09:13:00 2010] Checking reads
    seed length:     46bp
    format:         fastq
    quality scale:     phred33 (default)
[Fri Sep 17 09:13:46 2010] Mapping reads against hg18 with Bowtie
[Fri Sep 17 09:13:59 2010] Joining segment hits
[Fri Sep 17 09:13:59 2010] Searching for junctions via segment mapping
Warning: junction database is empty!
[Fri Sep 17 09:16:12 2010] Joining segment hits
[Fri Sep 17 09:16:12 2010] Reporting output tracks
    [FAILED]
Error: Report generation failed with err = 1

Any help would be appreciated!

[jdanderson]

Hello grahammcv,

I too have just experienced this problem. Have you found a resolution?

I noticed a thread on here entitled "Running ~35 bp and >=50 RNASeq reads" that might be a clue to a solution.

I am in the process of trying it out and will try to post results for everyone who might be having this same problem.

[zun]

Hi,everyone

I also got the same error as you all.
I doubted the index files or reads,but the cause was so simple in my case.

I saw the report.log and it says 'cant open BED file'.
I run tophat as root the other day,so junction.bed file was ownd by root.
That's why I got the error when running tophat as not root.

After changing owner of junction.bed , I got correct result.

Maybe you all don't make an elementary mistake like me
I would be glad this thread would help you.

[mgogol]

Oops, my mistake had to do with permissions on the annotation files as well.

[flobpf]

Because Tmp directory too big

I am getting this error too. I believe (atleast in my case) that it is due to disk getting full with the tmp directory. Errors I get are:

Reporting output tracks
        [FAILED]Report generation failed with err = -11

or

Error: Report generation failed with err = 1

Just before giving this error, tophat says:

sort: close failed: -: No space left on device

[edge]

Hi flobpf,

How you solve the problem about the sort error?
I'm facing the same error as well
Thanks for advice.

I am getting this error too. I believe (atleast in my case) that it is due to disk getting full with the tmp directory. Errors I get are:

Reporting output tracks
        [FAILED]Report generation failed with err = -11

or

Error: Report generation failed with err = 1

Just before giving this error, tophat says:

sort: close failed: -: No space left on device