Grouping fasta entries from different files based on reference/name

[gevielr]

Hi everyone,

I currently have 24 fasta files (for 24 species) with thousands of contigs. These files are contigs constructed from a reference transcriptome, so each file has contig numbers that are the same as contig numbers in the other files. For example:

File 1 (SpeciesA):

>contig_1
AAAAAAAAAAAAAA
>contig_2
TTTTTTTTTTTTTTTT
>contig_3
CCCCCCCCCCCCCC

File 2 (SpeciesB):

>contig_2
TTTTTTTTTTTTTTTT
>contig_3
CCCCCCCCCCCCCC

File 3 (SpeciesC):

>contig_1
AAAAAAAAAAAAAA
>contig_2
TTTTTTTTTTTTTTTT


I want to create files for each contig, so that instead of having 24 fasta files (one for each species), I have a file for each contig number like:

File 1 (contig1):

>contig_1_SpeciesA
AAAAAAAAAAAAAA
>contig_1_SpeciesC
AAAAAAAAAAAAAA

File 2 (contig2):

etc..


My programming skills are pretty much non-existent, so any direction on how I could do this for many large files would be highly appreciated!

[maubp]

My programming skills are pretty much non-existent, so any direction on how I could do this for many large files would be highly appreciated!

Learn to program. I would suggest Python, Ruby or Perl - have a look at online tutorials to see if which you like best, but also find out what people near you are using. Having someone able to help you out in person is really valuable, especially when first starting out.

[mikep]

Attached is a python script that will do what you want. you can invoke it on a mac or linux command line with

python splitter.txt directorywithoriginalcontigs

it will output the files you want in the directory you invoke the command. It assumes the species contigs sit in the above directory and are called
species1.fasta
species2.fasta

etc

Disclaimer: The above code is not guaranteed to work in the way intended, or indeed at all.