Not at all!

Thanks for the reply!

The above commands will take more time ( really so much of time ! )

Am testing with sample of 10 billion records.

So many process and kernel data structures will take days together to complete

I found the unix command to be useful.

cat filename | sort | uniq -c | sort -nr

What if I am using a tab-delimited text file with multiple columns?

Example:

S100A16 hsa-miR-125a-3p S100A16 0.0011959 0.768059
PBXIP1 hsa-miR-125a-3p PBXIP1 0.0199898 0.700326
CYB5R3 hsa-miR-125a-3p CYB5R3 0.0081174 0.748953
BEST3 hsa-miR-125a-3p BEST3 0.00148927 0.756234
FAM101A hsa-miR-125a-3p FAM101A 0.0196212 0.783555
KIAA0195 hsa-miR-125a-3p KIAA0195 0.0019755 0.747427
LLGL2 hsa-miR-125a-3p LLGL2 0.0248212 0.876563
FBLN5 hsa-miR-125a-3p FBLN5 0.0162988 0.776446
IFITM3 hsa-miR-125a-3p IFITM3 0.00896808 0.478704
SSH3 hsa-miR-125a-3p SSH3 0.0301693 0.836054
EXTERNAL_NAME SEQ EXTERNAL_NAME p-value(1 vs. 2) Ratio(1 vs. 2)
EMILIN1 hsa-miR-369-5p EMILIN1 0.0254294 0.720094
ADD3 hsa-miR-369-5p ADD3 0.0184075 0.742096
AIFM2 hsa-miR-369-5p AIFM2 0.00646348 0.829228
GPT2 hsa-miR-369-5p GPT2 0.00473291 0.706895

and I want the output to read

10 hsa-miR-125a-3p
4 hsa-miR-369-5p
1 SEQ

Initially, I made a new file, composing only the column of interest, using perl (column 1 here), named it filenam_list; then applied the

cat filenam_list | sort | uniq -c | sort -nr > filename_counts
rm filenam_list

Is there a more efficient way of doing this? I'm sure there has to be. I repeated the procedure on 7 files and I have to it again.

I think I found and answer to my own question. You specify the column of interest

If you don't have anything against (n|g)awk, simply do: