How many of the reads end up being counted by HTSeq count

Question

[Apex92]

I am new to these types of quality control checks. I have not faced any error in my analyses but want to perform quality control. I have aligned my sncRNA reads with bowtie (-m 1 option) and after making the bam output from the alignment I performed HTSeq count. Now I want to see if most of the reads actually end up being represented in the HTSeq file as gene counts. How can I do that?

Should I just simply sum up the total of the counts reported by HTSeq and divide it by the total number of unique reads?

Any help will be appreciated, thanks

Answer 1

[Ido Tamir]

https://htseq.readthedocs.io/en/master/count.html#usage at the bottom of the file you have some lines with some special counters you have to exclude for calculating the ratio. All of it has to sum up to the number of input reads. http://bioinf.wehi.edu.au/featureCounts/ is faster and gives you the same output, but the special counters are in a separate file.

[Apex92]

Thank you, Tamir for the answer. So after removing that extra lines, should I just simply sum up the total of the counts reported by HTSeq and divide it by the total number of unique reads?

[Ido Tamir]

better to sum up without the extra lines, sum up with the extra lines and then divide.