is "sequence Duplication levels" has bad effect on Alignment process?

Question

[modarzi]

Dear All,

I am studying Pairend RNA-seq data. In the first step, through the FastQC tool, I checked the quality of samples. After trimming, all measurements in the FastQC report are perfect except "sequence Duplication levels".Now, I need to align them to reference genome through "bwa" aligner. Considering paired study, I want to know if "sequence Duplication levels" can bad effect on my alignment?

I appreciate it if anybody shares his/her experience with me.

Best Regards,

[rpolicastro]

If this is RNA-seq for a eukaryote, you should use a splice aware aligner such as STAR instead of BWA. Better yet, if you just need to quantify feature counts, you can use a pseudo-aligner like Salmon.

[ATpoint]

OP is using some pipeline for circRNAs in which this is hardcoded C: should each RNA-seq has Circular RNA?.

[modarzi]

Thank you for your comment. I agree with you. But I need bwa SAM output for use as the input of CIRI2. CIRI2 is a CircularRNA identifier. In fact, bwa is not my choice, and using that is obligatory.

Answer 1

[ATpoint]

Considering paired study, I want to know if "sequence Duplication levels" can bad effect on my alignment?

No, duplication is normal and expected in RNA-seq.

[modarzi]

do you mean that in each report of FastQC relate to RNA-seq we have a red condition about "sequence Duplication levels"?

[ATpoint]

It is not unusual to have a read there, yes, but as it is expected you should not care about it. The only (for me) informative fastqc value (in RNA-seq) is adapter content and base quality, and the latter only if it is flagged indicating a technical problem during sequencing.