HT-seq Counts Stranded option

Question

[Krish Srinivasan]

Hi everyone,

I am having a trouble understanding HT-seq Counts. I have a first strand library (dUTP). This data was previously analyzed incorrectly using the option stranded = Yes by someone else. I have reanalyzed it with Stranded = Reverse which is the correct option. Here is a subset of the count table from one sample:

                        Gene    Stranded = Yes  Stranded = Reverse
          ENSDARG00000000001            7               114
          ENSDARG00000000002            12               37
          ENSDARG00000000018            28             1047
          ENSDARG00000000019            24             1236
          ENSDARG00000000068            8               308 
          ENSDARG00000000069            17             1014
          ENSDARG00000000086            25               88
          ENSDARG00000000103            98             4159
          ENSDARG00000000142            2                98
          ENSDARG00000000151            1                34

Can someone please help me understand how HT-seq count is doing the counting. Why does using stranded =Reverse with the data set produce more read counts than Stranded = Yes. I need to be able to understand the underlying reason why the counts are so different.

Thanks,

[o.mikh90]

Hi! Have you found the answer? Thank you!

[Krish Srinivasan]

Hi, please take a look at this for an explanation https://www.preprints.org/manuscript/201903.0124/v1

Answer 1

[GenoMax]

See @Devon (and other's) answers in these threads:
what is difference between strand specific in rnaseq analysis from htseq-count
HTSeq-count paired-end --stranded=yes gives less counts than --stranded=no
Taking into account antinsense reads when analyzing stranded Illumina data with DESeq2

[Krish Srinivasan]

thanks for the reply, but i have looked through those already and it didn't answer my question.