low alignment rate for RNA-Seq data

Question

[analyst]

Dear scientists,

I have RNA-Seq data of rice samples. I analyzed two rice datasets from two different labs (Fastp, Hisat2, Samtools, FeatureCounts). Alignment rate of one rice dataset is 98% on average however alignment rate of other rice dataset ranges between 26% to 63%. Both datasets were sequenced together at the same time on same run.

Ideally alignment rate should be above 80% but i can not discard the dataset with low alignment rate. How can i utilize it. Do i need to extract mapped reads and perform quantification on mapped reads only. Please guide me.

Thankyou so much!

[michael.ante]

Hi,

Check for rRNA reads, if they are still included in the read set (issues during rRNA depletion, or similar). These reads map usually to too many locations and are removed from the alignment. Other than that check for other sources of contamination. If you found it, you may want to remove the reads mapping to that source (e.g. using bbsplit).

[analyst]

Thankyou michael.ante! I will extract unmapped reads and blast to check contamination. I will let you know then. How to check for rRNA reads?

[GenoMax]

How to check for rRNA reads?

Align to rDNA repeat for Rice. Or use a tool like SortMeRNA: https://github.com/sortmerna/sortmerna