Counting single end reads

Question

[valopes]

Hi everyone,

I mapped my single end reads, with transcripts as reference, using Bowtie2. Now, I am trying to count both unique and multiple mapped reads. The multiple reads I need to count just for those transcripts that had at least one unique read mapped.

Does someone know how can I do this, please? Can I do it using HTseq?

Thanks

[WouterDeCoster]

Can I do it using HTseq?

Have you considered reading the manual? You can optionally include multimapping reads in your counts, but for the other requirement:

the multiple reads I need to count just for those transcripts that had at least one unique read mapped.

You will probably have to write a custom script.

I mapped my single end reads, with transcripts as reference, using Bowtie2.

I'm not sure what you are doing, but this is not the most appropriate method. Since you didn't specify the organism you are working on I am not sure if you have a reference genome, but most commonly you would use a splice aware aligner (e.g. STAR or HISAT2) to do the alignment on the genome.

[valopes]

Thank you, WouterDeCoster.

I will try to use STAR so.

[Carlo Yague]

the multiple reads I need to count just for those transcripts that had at least one unique read mapped.

A lazy way to do that would be to run HTseq twice :

• Once considering only uniquely mapped read to identify the transcripts with at least one unique read mapped.
• Once considering also multimappers to get the actual counts.