Error with Bowtie 2

Question

[valerie]

Hi guys,

I am working with exomes from TCGA which are bam files. I am converting bam to fastq using SamToFastq by Picard and then realign the reads back to reference genome using another reference, not like in original bam files. With one file I get an error from Bowtie2:

Error, fewer reads in file specified with -2 than in file specified with -1

However Bowtie2 doesn't fall, it goes on running forever. Does anybody know how to solve this issue? I do not get such an error for other files I worked with from TCGA. I suppose that these unpaired reads should be simply ignored by Bowtie2.

Thanks in advance!

[h.mon]

It seems R1 and R2 fastq files have different number of reads. Run fastqc or wc -l or zcat | wc -l to get number of reads for each file.

How were your picard and bowtie commands lines?

[valerie]

That's true, I see that the files have slightly different sizes. My script:

~/data/samtools-1.3.1/samtools sort -@ 64 -n filename.bam -o ~/data/tcga_results/filename_sorted.bam
java -jar ~/data/picard.jar SamToFastq I=~/data/tcga_results/filename_sorted.bam F=~/data/tcga_results/filename_R1.fastq 
F2=~/data/tcga_results/filename_R2.fastq
bowtie2 -x ~/data/Homo_sapiens/UCSC/hg38/Sequence/Bowtie2Index/genome -p 64 -1 ~/data/tcga_results/filename_R1.fastq 
-2 ~/data/tcga_results/filename_R2.fastq -S ~/data/tcga_results/filename_reassembled.sam

[GenoMax]

Use repair.sh from BBMap suite to re-sync the files before remapping them.

[valerie]

This worked! Thank you again!

[h.mon]

Try fixing your unpaired fastq files with BBTools:

repair.sh in1=r1.fq.gz in2=r2.fq.gz out1=fixed1.fq.gz out2=fixed2.fq.gz outsingle=singletons.fq.gz

I don't use much Picard, but there may be some magic flag to output properly paired fastq files.

[valerie]

This worked! Thank you!

[manojkumarbioinfo]

It worked fine thank you. the reads were not matching that's the problem