Dear all,
I want to use cutadapt when processing NGS data in order to remove 4 nucleotides at both 3-end and 5-end of each read. Could you suggest a command that I can use to achieve this purpose? Thank you for your help!
Hi Truc!
Here you go:
cutadapt -u 4 -u -4 -o trimmed.fastq reads.fastq
Where:
"cutadapt" is the installed program name
"-u" signifies cut, followed by 4 means the beginning bases and -4 means the end of reads
"-o" followed by filename to save
Then followed by the actual read file.
Hope it helps
Happy sciencing!
I'm assuming you are wanting to align a NGS sequencing readset against some targeted sequences or assembly - if not then please ignore the following comments.
Think about your alignment objectives and the underlying error profiles inherent in your readsets; to gain confidence in any individual read alignment you want a maximal read length K alignment combined with minimising the number of base mismatches allowed in that alignment. The error profile of Illumina (I'm assuming your reads are Illumina) is that the majority of sequencing errors are towards the 3' end of the reads, with partially retained adaptors sometimes an issue at the 5' end. If you have a budget in terms of the minimum read alignment K you can accept then you would be better off trimming more from the 3' end of the reads and less from the 5' ends.
In my previous life, I'm now retired :-), when aligning hard trimmed (with say 'cutadapt') 100bp reads against human then I used to advise trimming 5bp from the 5' end and 15bp from the 3' end and only allowing at most 5 mismatches.
Above considerations applies with either SE/PE reads, and irrespective as to if they are RNA-seq or WGS.
- Badges
- Science method