Some questions about the *.decontam.counts.tsv ? about clockwork HOT 4 CLOSED

After samtools view -F 4 ERR067664.sam | awk '$3=="NC_000962.3"' | wc -l is executed, the result is 2206287, which is consistent with the number of TB reads in the tsv file !
Thank you very much for your patient answer！

from clockwork.

Not sure right now why the difference of 1065. Clockwork just counts the number of reads matched to H37Rv, and that's what it reports in TB.

Your second question: this will be because of read pairs. If a read matches H37Rv, then the both reads of the pair are kept regardless of what the read's mate did (it could be unmapped or mapped to a contamination sequence). Either both reads of a pair are kept, or they are both removed. The surviving read pairs are what is written to the output fastq files - the total number of which will be the count in Reads_kept_after_remove_contam.

from clockwork.

Is my method of using samtools view -F 4 ERR067664.sam | grep NC_000962.3 | wc -l to obtain the number of reads matched to H37Rv consistent with your method? Is there something wrong with my statistical method?

Regarding the second question, I understand, thank you for your answer！

from clockwork.

grep NC_000962.3 will be looking for NC_000962.3 anywhere in the line, not just in column 3. This means it counts a read that is mapped to anything, if its mate is mapped to NC_000962.3.

Try awk '$3=="NC_000962.3"' instead of grep NC_000962.3

from clockwork.