Hello, I have sequence data from the Illumina platform. It is paired-end sequence data. I want to cut the adaptors for the same. I also have QC reports for each sample in fastq format. some of the sample reports show the green tick for the overrepresented sequences and yellow for adaptor content, whereas some show a yellow exclamatory symbol on overrepresented sequences and green for adaptor content. how do to chose which samples to trim for adaptors? should I trim only samples that show overrepresentation? or should I apply trimming to all samples irrespective of QC?
And if we had to choose some threshold for trimming adaptors what is best for DGE using RNA seq?
For general RNA-seq analysis you don't need to trim adapters, in fact it's recommended against doing this. They will be soft clipped during alignment so don't pose a problem.
Okay, so if I understood right, direct alignment can be performed after QC. But when you say general RNA seq analysis, what type of RNA seq analysis will need trimming adaptors? Sorry if the question sounds naive, am new to the field of computational biology. Thank you so much for the response.
By general RNA-seq I mean most analysis where feature level quantification is all you care about, i.e. differential expression. I would only trim adapters for tasks that are sensitive to the exact transcript structure, like de novo transcript assembly.
Thank you for the response.
Since it's RNA-seq data, the overrepresented sequences are something that tolerable because the RNA-seq data which were generated usually obtain from RNA samples that do not subject to random fragmentation and repeats are common in RNA molecules. You can trim the sample data that showed up warning by adaptor content module by cross-referring to universal adaptors libraries that are in-built in your selected trimming software and re-check the data for improvement by re-run the QC.
Thank you so much for your response.
Log in to answer this question.
