Trimmomatic on RNAseq data

It is possible to consider different qualities of average for the samples of a study in the trimmomatic?

Hello @Fateme

Do you mean is the Trimmomatic tool configurable? If so, yes, the tool form has parameters that can be adjusted by the user.

There are also other QA/trimming tools. Some may have more options that others.

Commonly used tools have examples in our tutorials. Start here Search GTN Materials (query=quality).

This is a screenshot of the tool form with the defaults. Each of these values can be adjusted. And, the pull-down menus have more choices, plus you can combined multiple filtering choices together.

The best advice is to experiment! Your original data is preserved, and you won’t “break Galaxy” even if the parameters are not great. The worst is a failed job, and the logs might be informative enough to help you decide what to try next. :slight_smile:

Hello, thank you for reply. What I mean is, is it possible to trim a sample from quality 20 in one study and use a higher quality (for example 22) for another sample of the study? Or should all the samples of a study be trimmed with only one quality average (for example 22) ? According to the screenshot that I send you. thank you.

Hi @Fateme Thanks for explaining.

Technically, yes, you can filter different samples differently.

Scientifically, there should probably be a good reason, and you might need to watch out for introducing some new bias into the data by accident. And would want to explain that to any readers of your results.

I can’t think of a good reason off hand to do this differently, especially based on quality scores, from samples all from the same study. Meaning, if you have some minimum quality threshold you want to apply to reads, why would some data need a different threshold pre-mapping? The next steps also have minimum thresholds for “hit” versus “no-hit”, and I wouldn’t recommend adjusting that differently per-sample.

So – what you are asking about will probably not make a difference, but could, which loops back around to having a good reason to do it all all :slight_smile: