creating tag count file

Question

[Ric]

Hi, I used FASTP but it appears that it does not create tag count. On the other hand, preprocess.seq is able to produce it.

It appears that tag-count format is normalized and has the following format:

$ head 2600_norm.txt
==> 2600_norm.txt <==
AAAAAAAAAAAAAAAAAAA 1
AAAAAAAAAAAAAAAAAAAAAA  1
AAAAAAAAAAAAAAAAAAAAAAA 1
AAAAAAAAAAAAAAAAAAAAAAAA    2
AAAAAAAAAAAAAAAAAAAAAAAAAA  1
AAAAAAAAAAAAAAAACTTGCGGCC   1
AAAAAAAAAAAAAAAATGTCTGGACA  1
AAAAAAAAAAAAAAACCCTAGCCT    1
AAAAAAAAAAAAAAATCAATGGCGTA  1
AAAAAAAAAAAAAAATCGCGGCC 1

Is there a way to create tag count from FASTP results?

Thank you in advance,

[Devon Ryan]

We don't normally use the word "tag" in this context anymore (that hasn't been used much in the past ~8 years). What is your end goal with this? For what it's worth, fastp is meant for read trimming and QC, nothing else.

[Ric]

I want to use PHASIS, and one of the supported formats is a tag-count. However, it was not working with the FASTA file, but I heard that it worked with tag-count.

[Devon Ryan]

It's unfortunate that they don't document what "tag-count" format is, since it's not something commonly used or that I've even heard of.

[Ric]

I found the tag-count files here and it appears that they did sort normalization. However, I do not how to create this file.

Answer 1

[GenoMax]

If those are simple counts of sequences then you could generate them by using reformat.sh from BBMap suite :

reformat.sh in=your.fq out=stdout.fa | grep -v "^>" | sort | uniq -c > sequence_count

[Ric]

I have got 5 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA but I need AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA 5. How is it possible to change the above command?

[Ric]

Thank you, I added awk ' { t = $1; $1 = $2; $2 = t; print; } ' input_file and it works.