script suggestion (SAM file)

[Sergio.pv]

Hi everyone!
In this attached example, i would like to know how many times the element
"bta-miR-191" and "bta-miR-10" in the header of a SAM file is repeated in the rest of
the document (in this case: 6 and 5 respectively).

Could you give me an idea for an script based on this example?

Thanks

[shoegame2001]

grep -v '@SQ' example.txt | awk '{if ($3=="bta-miR-191") print}' | wc -l


grep -v '@SQ' example.txt | awk '{if ($3=="bta-miR-10") print}' | wc -l

[dariober]

Hello- To expand a bit shoegame2001's answer... If your input file is big you might want to read thorough it only once and count the two (or more) patterns at the same time:

grep -v '@SQ' example.txt \
| awk '{if ($3=="bta-miR-191")
            mir191+=1;
        else if ($3 == "bta-miR-10")
            mir10+=1}
        END {print "bta-miR-191:" mir191, "\nbta-miR-10:", mir10}'

Dario

[Sergio.pv]

thanks, the grep function was very usufull to remove the headers.
After doing that, i did a count ot he repited elements on column 3.
cat myfile-noheaders.out | awk '{print$3}' | sort | uniq -c | sort -rnk1 >SAM-counts.out

I hope this can be helpfull for somebody else!

[Sergio.pv]

if it is usefull, here the complete script to remove the headers ('@') and then count repeated terms in column 3 of the SAM file:

grep -v '@' myfile.out | awk '{print$3}' | sort | uniq -c | sort -rnk1 >SAM-counts.out

Cheeers