Trimming DNAStringSet

Question

[a.krassnig]

Hello,

I am currently dealing with the problem of reading in a Fastq-File with "readDNAStringset", trimming the Sequences and then writing them in to a new fastq-file.

The reading of the fastq-file with "readDNAStringSet" is working just fine.

I am then trying to trim a fixed length from the left side of the Sequence (ex.10 Bases). Right now I use "subseq(my_Stringset, start =10)".

But when I try to write the fastq-file it seems like I have not trimmed the QualityScores and get the Error message "'x' and 'quality' must have the same width".

I have been looking for a while now, how to trim the sequence aswell as the Q-Score, but I just can't find a Solution (Which should be also fast, since I am working with very big Files).

Is there an correct way to do, what I am trying to do?

Thank you in advance.

I am using R 4.1.0

[cpad0112]

It's working fine here (R 4.0, Biostrings 2.58.0). Please post example fastq.

[a.krassnig]

Hey, thank you for your answer,

As far as I can see the ReadDNAStringset only saves the qualities if you set the parameter "with.qualities = TRUE". Otherwise you dont save the Quality-Values at all and also dont save them to your fastq-file. Am I correct here?

I think that is why the Qualty is a ";" for all bases. I just tried it and it also does that with my example.

Have a great day :)

Answer 1

[Michael]

I am not sure about the level of support for fastq in Biostrings. Please use the ShortRead package instead and use its built-in trimming functions. Are you sure you want to trim a fixed size off from the reads? This shouldn't be required if the reads were processed properly. I suspect this could be motivated by a misinterpretation of fastQC reports. Normally, one would only trim in case there is adapter content or low quality in a window, and otherwise leave the reads alone.

[a.krassnig]

Dear Michael,

Thank you for your answer. It actually is not the same trim length for every Sequence. I just thought this would be mor easy for my question. Actually I go through the Sequences an Trim depending on a calculation and allignment to another Sequence.

I have found ShortRead but was not sure if I should use it instead of Biostring. Is it possible to read the Sequence with Biostring and manipulate it with ShortRead?

[Michael]

I haven't tried this yet, and I don't think you can convert the objects between the packages but should be easy to test.