Hello, I'm new with rnaseq and I'm having trouble deciding if I should trim or not and if yes, which tool (trimmomatic or trimgalore). I have rnaseq with mixed reads from mammalian host and bacteria, and am planning to use Trinity for de novo assembly of microbiomes. However, I need to remove the host first. Sequences are very good quality but have some adaptors and a lot of ribosomals despite the use of sortmerna. I have 2x150 pair reads. I was trying to use trimmomatic but I can't decide which parameter of minlen to use without loosing too much information, and also parameters of sliding window 4:30 but why not 10:30? or illuminaclip ILLUMINACLIP:TruSeq3-SE.fa:2:30:10 , in which cases should I change it to 1:30:1 for example. I know there are a lot of tools for trimming, and overall I can't decide if I should even trim these sequences or removing the adapters.
1 answer
planning to use Trinity for de novo assembly of microbiomes
Any time de novo work is planned, you want to make certain that there is no extraneous sequence left. In your case before assembly (this will include adapters etc along with the host).
However, I need to remove the host first.
That can be the priority then.
You could use a tool like bbsplit.sh from BBMap suite to bin the reads (use known host to remove/bin those sequences away from others). bbduk.sh would then be used for cleaning/trimming the reads. More info about the tools here:
bbsplit: https://bbmap.org/tools/bbsplit
bbduk: https://bbmap.org/tools/bbduk
Log in to answer this question.