Entering edit mode
5.7 years ago
star
▴
350
I have run Bowtie2 for aligning ChIP-seq data and in one data I faced the below error
Error: Read SN7001318:103:H86L8ADXX:2:2102:16521:69527 1:N:0:8 has more read characters than quality values.
terminate called after throwing an instance of 'int'
Aborted (core dumped)
(ERR): bowtie2-align exited with value 134
I like to know, can I continue with the output of bowtie (I still have .bam file) or I should remove related read?
In my experience that happens when you are out of memory. Should check memory consumption first. Next,
grep
out that read:zgrep 'SN7001318:103:H86L8ADXX:2:2102:16521:69527' your.fastq.gz
and check if it is malformatted.Yes, I have enough, at least for a .bam file.
But perhaps not enough for the file to be written completely and then have the process exit normally? As of now the aligner is crashing without a normal program ending.
Note: Assuming the read checks out as correct/normal using @ATPoint's command. Add
-A 3
to that command to get the full fastq record for that ID.Thanks for reply.
I did it and I think it does not have enough quality score.
Shall i ignore it?
Try running
repair.sh
from bbmap and then check how many reads have been eliminated. If it is a minor number, just ignore it.It is only one read (last read). Would you please let me know , How Can I remove it?
That may be an indication that even the trimming may have not been complete and resulted in an incomplete fastq record?
I still would re-do the trimming and pipe it directly to bowtie2 like (given your trimmer supports writing to
stdout
,cutadapt
does)