Ovation RNA-seq system, bionfomatic alignment and htseq-count

Question

[nanoide]

Hello, We have used ovation RNA-seq system and prepared some libraries. We have sequenced single-ends. We want to use hisat2 and htseq counts to get counts per gene in our organism. I wondered if someone has worked with this before, what parameter should be specificied in rna-strandness on HISAT2 parameters.

Also,I understand that parameter determines whether a read correspond to a transcript, or to the reverse complemented counterpart of a transcript. But afterwards, in htseq-count, the stranded parameter depends also on this? I mean, if we use --rna-strandness F in hisat2, then we have to use --stranded yes or --stranded reverse in htseq-count?

Thanks for your help

Answer 1

[caggtaagtat]

Hi, it depends on how you build your library. If you don't have a stranded library, you don't need to set this parameter.

If you have a stranded library, there are two options for you in HISAT2 (This information can be found in the sequencing protocol, if the library preparation was done extern). Either your reads represent the sequence of the RNA segements, which you sequenced. Then your library is forward stranded and you have to set the parameter to --rna-strandness F.

If your reads represent the reverse complemented counterpart of your RNA fragments, which you sequenced, your library is reverse stranded and you have to set the strandness parameter to --rna-strandness R.

In htseq-count, the --stranded yes would be the equivalent to --rna-strandness F in HISAT2

[nanoide]

Thank you so much for your clarifying response. I have inherited some code where they first mapped using hisat2 and --rna-strandness F and then performed conts using htseq-count and --stranded reverse. Then this combination is not correct right? Thanks for your time

[caggtaagtat]

Glad I coudl help. Yes that sounds wrong to me. The strandness of a given data set remains the same, as long as the sequence of the reads is not overwritten by its reverse complemented counterpart in your inherited code, for whatever reason.

[GenoMax]

There is more than one type of Ovation kits. Not sure if all of them are stranded. So be sure to check what kit was used for this data.

[nanoide]

Will do. Thank you both