Data management in R, perl or python

Question

[MSRS]

Hi, thank you for answering my problem.

My data is in format as below:

A1P
D4M
N6G
A1F
D4S
N6L
A1C

I want the output or output should be :

A1P/F/C
D4M/S
N6G/L

Is any R package code available? perl or python code will also be great. Thank you very much. Sorry for wasting your valuable time.

[Arup Ghosh]

Please add more details to your question. What does "/F/C" and "\L" stand for?

[Ido Tamir]

they want the suffix after the 2 letters collected and appended. This is a basic programming question for which they should show at least some effort in some language

[MSRS]

English Alphabet! Basically, it will be used for amino acid (single letter code) and Nucleotide data formating.

[Ram]

I'd look out for residue positions >9 - that will result in total length being >3, and scripts below that don't account for it will fail. JC's solution will work best in that case.

[JC]

yeah, I was thinking the OP could have a position >9 in the inputs

Answer 1

[JC]

Perl:

#!/usr/bin/perl
use strict;
use warnings;
my %data;
while (<>) {
    chomp;
    if (m/(\w\d+)(\w)/) {
        my $key = $1;
        my $new = $2;
        if (defined $data{$key}) {
            $data{$key} .= "/$new";
        }
        else {
            $data{$key} = $new;
        }
    }
}
while (my ($key, $aa) = each %data) {
    print "$key$aa\n";
}

Test:

$ perl comb.pl < list.txt
A1P/F/C
D4M/S
N6G/L

[MSRS]

Thank you, JC. Excellent!

Answer 2

[mohammadhassanj]

python solution

from collections import defaultdict
result = defaultdict(str)
for line in open("input.txt").readlines():
  line = line.strip()
  result[line[:2]] = "/".join([result[line[:2]],line[-1]])
with open("output.txt","a") as file:
  for first,second in result.items():
    file.write(first+second[1:]+"\n")

[MSRS]

Thank you for sharing your scripts.

[Ram]

By the way, you don't need to bookmark every answer. You can bookmark the top level post, and that way you'll have access to all the answers.

[MSRS]

Sorry for that. I will follow your instruction. Thank you very much for the correction.

[Ram]

Don't worry about it - it's not a "Don't do this", it's just "you don't need to". Our bookmarks section can get cluttered easily.

Answer 3

[Pierre Lindenbaum]

sed 's/^\(..\)/\1\t/' input.txt | datamash  -t $'\t' -s -g 1  collapse 2 
A1  P,F,C
D4  M,S
N6  G,L