names of R1/R2

[nservant]

it seems that sometimes the read names in R1/R2 files are not exactly the same.
if it happens, HiCPro currently launch an error and stops ...

[milkcookie]

Did you have any suggestion to solve it? Or Which version can avoid it?

[nservant]

I did not fix it yet.
Could you first check it by looking at the first raw of your fastq files.
And if that's the case, I would suggest to modify the read name of your fastq files.

[milkcookie]

ok , here is the name of two sequence
@E00492:288:HJLVLCCXY:2:1101:2991:1854 1:N:0:GGCTAC
@E00492:288:HJLVLCCXY:2:1101:2991:1854 2:N:0:GGCTAC
it has '1' and '2' in it.

[milkcookie]

in merged bam file here is the name
E00492:288:HJLVLCCXY:2:1101:1763:5282
E00492:288:HJLVLCCXY:2:1101:1763:5282
I think is ok
but I run successful in recent days before with the data, now its failed with a new assembled genome.

[nservant]

indeed.
I would suggest to first save your files somewhere :)
Then something like
perl -pi -e 's/2:N/1:N/' will transform all 2:N into 1:N ...
Or simply remove the last part starting with 1:N ...
N

[nservant]

btw, can I ask you why type of sequencer it was ??
because so far, I do not know when the sequencer add this flag or not ...

[milkcookie]

the sam merge terminate after run part of two sam file, becouse of this problem.

[milkcookie]

it seq by company, to assemble the genome.

[nservant]

yes because the merge expect to have the R1 and R2 files in the same order.
So to do that, it checks if the read names are the same ... which is not the case here.

[milkcookie]

I think is novaseq or hiseq2500. I don't know clearly.
I can run successful with another data seq by same company, and the same format....
I suggest you to add some code to skip the error line in bam.

[milkcookie]

Hi
I found the problem, its becouse the reads num in two fq file is diff. any suggestion to solve it?

[milkcookie]

Hi:
I try some test today, but when the fastq file has diff num of reads there will meet this problem?
how to avoid this diff happen in bam file by bowtie2? if I can add some args to skip?

[nservant]

what do you mean. The fastq files must have the same number of lines as this is PE sequencing.
Then, you just have to fix the issue with the read name. It should not affect the number of lines in you fastq ?

[nservant]

Hi,
I think I fixed it.
In order to test it, would you mind sending me two small SAM files with a few reads ?
Thanks

[milkcookie]

Hi
I have some PE clean data with diff num of reads.
I don't know why it happen?
can you give me some suggestion?
the problem happle becouse of the diff num of reads, not the seq name?

[nservant]

Difficult to say. What do you mean by 'clean' ? PE raw data should always be paired with same number of reads in the two files. Could you have access to the files before cleaning ?

[milkcookie]

yes
I have raw data
the clean data had been trimming adapter and dumping the low quality reads, I don't know what happened, the company do it.
so I want to know is there some tools to correct the reads, so it can be same numbers in R1 and R2.

[nservant]

should be fixed now in devel (v2.11.0)