tophat2 Error: segment-based junction search failed with err =-9

Question

[marongiu.luigi]

Dear all, I am aligning HISeq RNA-Seq data to the human reference genome. I downloaded the Homo_sapiens.GRCh38.dna.toplevel.fa.gz and the Homo_sapiens.GRCh38.84.gtf.gz reference files and I built the reference indices with the following:

bowtie2-build -f GRCh38.84.fa GRCh38.84
tophat2 -p 16 -G GRCh38.84.gtf --transcriptome-index=GRCh38.84.tr GRCh38.84

which gave me the following files: GRCh38.84.1.bt2l GRCh38.84.2.bt2l GRCh38.84.rev.1.bt2l GRCh38.84.3.bt2l GRCh38.84.rev.2.bt2l GRCh38.84.4.bt2l

and the folder GRCh38.84.tr which contains: GRCh38.84.1.bt2 GRCh38.84.4.bt2 GRCh38.84.gff GRCh38.84.ver GRCh38.84.2.bt2 GRCh38.84.fa GRCh38.84.rev.1.bt2 GRCh38.84.3.bt2 GRCh38.84.fa.tlst GRCh38.84.rev.2.bt2

I have two paired files seq_1.fastq and seq_2.fastq, I removed the TruSeq adapters with trimmomatic with:

java -jar /usr/bin/trimmomatic.jar PE -threads 16 -phred33 seq_1.fastq seq_2.fastq seq_1Paired seq_1Unpaired seq_2Paired seq_2Unpaired ILLUMINACLIP:./TruSeq.fa:2:30:10:1:true

then I ran the alignment with Tophat:

$ tophat2 -o outputFolder -G GRCh38.84.gtf --transcriptome-index=GRCh38.84.tr --no-coverage-search -p 16 GRCh38.84 seq_1Paired seq_2Paired

and I got this error -9 with the log:

[2016-05-15 08:08:59] Beginning TopHat run (v2.1.1)
-----------------------------------------------
[2016-05-15 08:08:59] Checking for Bowtie
          Bowtie version:    2.2.6.0
[2016-05-15 08:09:00] Checking for Bowtie index files (transcriptome)..
[2016-05-15 08:09:00] Checking for Bowtie index files (genome)..
[2016-05-15 08:09:00] Checking for reference FASTA file
[2016-05-15 08:09:00] Generating SAM header for GRCh38.84
[2016-05-15 08:11:02] Reading known junctions from GTF file
[2016-05-15 08:12:32] Preparing reads
     left reads: min. length=12, max. length=60, 26138268 kept reads (2188 discarded)
    right reads: min. length=12, max. length=60, 25981749 kept reads (158707 discarded)
Warning: short reads (<20bp) will make TopHat quite slow and take large amount of memory because they are likely to be mapped in too many places
[2016-05-15 08:42:52] Using pre-built transcriptome data..
[2016-05-15 08:43:11] Mapping left_kept_reads to transcriptome GRCh38.84 with Bowtie2 
[2016-05-15 12:56:28] Mapping right_kept_reads to transcriptome GRCh38.84 with Bowtie2 
[2016-05-15 17:28:33] Resuming TopHat pipeline with unmapped reads
[2016-05-15 17:28:33] Mapping left_kept_reads.m2g_um to genome GRCh38.84 with Bowtie2 
[2016-05-15 18:33:01] Mapping left_kept_reads.m2g_um_seg1 to genome GRCh38.84 with Bowtie2 (1/2)
[2016-05-15 18:40:20] Mapping left_kept_reads.m2g_um_seg2 to genome GRCh38.84 with Bowtie2 (2/2)
[2016-05-15 18:51:34] Mapping right_kept_reads.m2g_um to genome GRCh38.84 with Bowtie2 
[2016-05-15 19:53:43] Mapping right_kept_reads.m2g_um_seg1 to genome GRCh38.84 with Bowtie2 (1/2)
[2016-05-15 20:01:00] Mapping right_kept_reads.m2g_um_seg2 to genome GRCh38.84 with Bowtie2 (2/2)
[2016-05-15 20:13:22] Searching for junctions via segment mapping
    [FAILED]
Error: segment-based junction search failed with err =-9
Loading left segment hits...

The files I obtained are in the folders logs and tmp: ./logs$ ls bowtie.left_kept_reads.log m2g_left_kept_reads.err bowtie.left_kept_reads.m2g_um.log m2g_left_kept_reads.out bowtie.left_kept_reads.m2g_um_seg1.log m2g_right_kept_reads.err bowtie.left_kept_reads.m2g_um_seg2.log m2g_right_kept_reads.out bowtie.right_kept_reads.log prep_reads.log bowtie.right_kept_reads.m2g_um.log run.log bowtie.right_kept_reads.m2g_um_seg1.log segment_juncs.log bowtie.right_kept_reads.m2g_um_seg2.log tophat.log gtf_juncs.log

./tmp$ ls GRCh38.84.bwt.samheader.sam GRCh38.84_genome.bwt.samheader.sam GRCh38.juncs left_kept_reads.bam left_kept_reads.bam.index left_kept_reads.m2g.bam left_kept_reads.m2g.bam.index left_kept_reads.m2g_um.bam left_kept_reads.m2g_um.bam.index left_kept_reads.m2g_um.mapped.bam left_kept_reads.m2g_um.mapped.bam.index left_kept_reads.m2g_um_seg1.bam left_kept_reads.m2g_um_seg1.bam.index left_kept_reads.m2g_um_seg1.fq.z left_kept_reads.m2g_um_seg1_unmapped.bam left_kept_reads.m2g_um_seg1_unmapped.bam.index left_kept_reads.m2g_um_seg2.bam left_kept_reads.m2g_um_seg2.bam.index left_kept_reads.m2g_um_seg2.fq.z left_kept_reads.m2g_um_seg2_unmapped.bam left_kept_reads.m2g_um_seg2_unmapped.bam.index left_kept_reads.m2g_um_unmapped.bam left_kept_reads.m2g_um_unmapped.bam.index right_kept_reads.bam right_kept_reads.bam.index right_kept_reads.m2g.bam right_kept_reads.m2g.bam.index right_kept_reads.m2g_um.bam right_kept_reads.m2g_um.bam.index right_kept_reads.m2g_um.mapped.bam right_kept_reads.m2g_um.mapped.bam.index right_kept_reads.m2g_um_seg1.bam right_kept_reads.m2g_um_seg1.bam.index right_kept_reads.m2g_um_seg1.fq.z right_kept_reads.m2g_um_seg1_unmapped.bam right_kept_reads.m2g_um_seg1_unmapped.bam.index right_kept_reads.m2g_um_seg2.bam right_kept_reads.m2g_um_seg2.bam.index right_kept_reads.m2g_um_seg2.fq.z right_kept_reads.m2g_um_seg2_unmapped.bam right_kept_reads.m2g_um_seg2_unmapped.bam.index right_kept_reads.m2g_um_unmapped.bam right_kept_reads.m2g_um_unmapped.bam.index segment.deletions segment.fusions segment.insertions segment.juncs temp.samheader.sam

Could you please tell me what I got wrong and how can I fix it?

Many thanks,

Luigi

[NK]

Dear Luigi, I get the same error. Have you made any progress with understanding what leads to it? Thanks, N

[rborgesm]

I also obtained this error on one of my files, were you able to figure out why this was?

[marongiu.luigi]

nope, I did not solve it...

Answer 1

[jesselee516]

It may be caused by running out of memory.. You can try one cell line each time with multiple processor, instead of running multiple cell lines at the same time.