Hi all,
I have a PE dataset which i need to filter so that I only keep those reads which have a specific sequence in exactly one of the two mates.
I.e. i want to discard all reads that
I tried cutadapt with it's --discard-untrimmed flag. However, it seems that cutadapts paired-end implementation, which currently requires two passes over the data, is not suitable since the first pass, over Read 1, already discards all the reads with no matching sequence => Read 2 of these discarded mates are never actually looked at.
The command I am using is from the cutadapt documentation:
Am I using cutadapt the wrong way or are there other tools that can do what I need?
thanks,
-dexter
I have a PE dataset which i need to filter so that I only keep those reads which have a specific sequence in exactly one of the two mates.
I.e. i want to discard all reads that
- don't have the sequence in either mate
- have the sequence in both mates
I tried cutadapt with it's --discard-untrimmed flag. However, it seems that cutadapts paired-end implementation, which currently requires two passes over the data, is not suitable since the first pass, over Read 1, already discards all the reads with no matching sequence => Read 2 of these discarded mates are never actually looked at.
The command I am using is from the cutadapt documentation:
Code:
cutadapt -a SEQUENCE_FWD --discard-untrimmed --paired-output tmp.2.fastq -o tmp.1.fastq reads.1.fastq reads.2.fastq cutadapt -a SEQUENCE_REV --discard-untrimmed --paired-output trimmed.1.fastq -o trimmed.2.fastq tmp.2.fastq tmp.1.fastq
thanks,
-dexter
Comment