error in bowtie2 alignment

Question

[prasoonagar]

Hi I have done the trimming of my reads using cutadapt after trimming i did FASTQC and find that three modules

• Per Base Sequence Content
• Over represented sequences
• Sequence Length Distribution
• Per Sequence GC Content
• Kmer Content

they all failed. But I still went ahead and tried aligning the reads using bowtie2 with the following command

bowtie2 -p 8 -x ../hg38_index/index -X 300 --no-unal --time --no-mixed --local --dovetail --very-sensitive -1 Lmin6-ChIP-41650664/trim_cutadapt/trim_Lmin6_chip_R1.fastq.gz -2 Lmin6-ChIP-41650664/trim_cutadapt/trim_Lmin6_chip_R2.fastq.gz -S Lmin6-ChIP-41650664/Lmin6_chip.sam

I got the following error with the bowtie2

Warning: minimum score function gave negative number in --local mode for mate #2 of read 'M00546:70:000000000-ANJWB:1:1107:18525:13236 2:N:0:CGATGT; setting to 0 instead
Warning: skipping mate #1 of read 'M00546:70:000000000-ANJWB:1:1107:18525:13236 1:N:0:CGATGT' because length (0) <= # seed mismatches (0)
Warning: skipping mate #2 of read 'M00546:70:000000000-ANJWB:1:1107:18525:13236 2:N:0:CGATGT' because length (0) <= # seed mismatches (0)
Warning: skipping mate #1 of read 'M00546:70:000000000-ANJWB:1:1107:18525:13236 1:N:0:CGATGT' because it was < 2 characters long
Warning: skipping mate #2 of read 'M00546:70:000000000-ANJWB:1:1107:18525:13236 2:N:0:CGATGT' because it was < 2 characters long
Warning: minimum score function gave negative number in --local mode for mate #1 of read 'M00546:70:000000000-ANJWB:1:1107:16473:13334 1:N:0:CGATGT; setting to 0 instead
Warning: minimum score function gave negative number in --local mode for mate #2 of read 'M00546:70:000000000-ANJWB:1:1107:16473:13334 2:N:0:CGATGT; setting to 0 instead
Warning: skipping mate #1 of read 'M00546:70:000000000-ANJWB:1:1107:16473:13334 1:N:0:CGATGT' because length (0) <= # seed mismatches (0)
Warning: skipping mate #2 of read 'M00546:70:000000000-ANJWB:1:1107:16473:13334 2:N:0:CGATGT' because length (0) <= # seed mismatches (0)
Warning: skipping mate #1 of read 'M00546:70:000000000-ANJWB:1:1107:16473:13334 1:N:0:CGATGT' because it was < 2 characters long
Warning: skipping mate #2 of read 'M00546:70:000000000-ANJWB:1:1107:16473:13334 2:N:0:CGATGT' because it was < 2 characters long
Warning: minimum score function gave negative number in --local mode for mate #1 of read 'M00546:70:000000000-ANJWB:1:1107:14918:13429 1:N:0:CGATGT; setting to 0 instead
Warning: minimum score function gave negative number in --local mode for mate #2 of read 'M00546:70:000000000-ANJWB:1:1107:14918:13429 2:N:0:CGATGT; setting to 0 instead

I tried another thing did the alignment using the untrimmed files and then the command worked perfectly. so want some suggestion if I did any thing wrong in the trimming. I used the following command for trimming

~/.local/bin/cutadapt -a AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC -A AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGTAGATCTCGGTGGTCGCCGTATCATT \
-o Lmin6-ChIP-41650664/trim_cutadapt/trim_Lmin6_chip_R1.fastq.gz \
-p Lmin6-ChIP-41650664/trim_cutadapt/trim_Lmin6_chip_R2.fastq.gz Lmin6-ChIP-41650664/Lmin6-ChIP_S1_L001_R1_001.fastq.gz   Lmin6-ChIP-41650664/Lmin6-ChIP_S1_L001_R2_001.fastq.gz

I will be happy if some kind suggestions are there for this problem.

Thanks

Answer 1

[Brian Bushnell]

I don't see a problem. You got warnings, not errors. Probably, it's because you have some very short reads after trimming; you can remove those if you want, which might make the warnings go away. With the BBMap package, you would do this:

reformat.sh in=Lmin6-ChIP-41650664/trim_cutadapt/trim_Lmin6_chip_R#.fastq.gz out=filtered#.fq.gz minlen=30

[i.sudbery]

cutadapt can be run in paired end mode and told to drop pairs where one read is shorter than a certain length. I generally drop anything shorter than 15 or 20bp

[prasoonagar]

Thanks for your suggestions i think you are right but one more query the warning says that skipping mate means they were already dropped during the alignment???

[Brian Bushnell]

It looks like those are length-0 reads. Not much there to align :)

[pdhrati02]

Hi Brian, I had a similar warnings for many many reads in my log file: Warning: skipping read #### because length (1) <= # seed mismatches (0) Warning: skipping read #### because it was < 2 characters long. I was using kneaddata with bowtie2.

My question is that how to find out if I really have many short reads and if the number is high how to tackle the issue?

Hope you can help me with this.

Looking forward to your reply. thank you

[i.sudbery]

Dear pdhrati02 Given the age of this question and the slight difference in your question, I think you would be better off starting a new question.