adding header using grep and word count in bash

Question

Hi everyone I have a lot (200 or so) files that look like this

>Sample_NameA
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
>Sample_NameB
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT

I need to add a header on every file that consist on the number of samples, the number of letters (A,C,G,T) and the number one.

So for example this first file should look like this:

2 441 1
>Sample_NameA
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
>Sample_NameB
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT

I was thinking of using first grep ">" to count the number of samples because every sample name is preceded by that and using the wc to count the number of characters until another >, but I couldn't manage it. In a moment I will upload my script with what I have so far, but if someone has resolved it jet I would really appreciated it.

Thanks id advance.

Answer 1

This awk can do all this in one command:

awk '{if (d) d=d RS $0; else d=$0}
      />/{s++;next}
      s==1 && /[ACGT]/{gsub(/[^ACGT]+/, ""); n+=length($0)}
      END{print s, n, "1" RS d}' file
2 441 1
>Sample_NameA
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
>Sample_NameB
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT
ACGTGTCTAGTATGCATGCATAGCTACGACTACGACTAGCATCAGCATCGACTAGCATCGACT

Answer 2

#!/bin/bash
DATA="$1" # for readability; give /path/to/data/file as first argument to the script
SAMPLES=$(grep -c '>' "$DATA")
CHARS=$(grep -v '>' "$DATA" | wc -c)
echo "$SAMPLES $CHARS 1"
cat "$DATA"

Answer 3

Can't think of a way to do this with just one grep, but certainly doable with a few pipes.

e.g.

> grep -E "^>" file 
2
> grep -E -v "^>" file | grep -o "[ACGT]" | wc -l
756

Answer 4

With full credit to @anubhava, here's a method that doesn't keep the entire file in memory:

ed file <<END
1i
$(awk '/^>/ {s++; next} s==1 {l+=length()} END {print s,l,1}' file)
.
w
q
END