Concatenating Sequences Within a single Fasta File

[cdlam]

Hi all,
I need to concatenate a bunch of sequences in a FASTA file. I have a file of extracted introns and would like to essentially splice them all together for use in a program. Is there any way to do either of these using perl (preferably) or python (if necessary):

1. Join all the introns of a single gene, preserving the FASTA heading for that gene.

> Gene 1 Intron 1
GTACGCC....CTGATAGAG
>Gene 1 Intron 2
GTCCAGGAC.....CTGAGTAAG

Becomes
> Gene 1 Intron 1
GTACGCC....CTGATAGAGGTCCAGGAC.....CTGAGTAAG

or
2. Join a number of introns together (not accounting for what genes they came from) under a non-specific FASTA formatted heading?

Basically I want to splice together a bunch of intron sequences like they were exons so that I can run them through a program that doesn't like how short they are. The first way would be the most biologically relevant and useful for my purposes, but if it can't be done I can live with it haha. Any help would be greatly appreciated. Thanks a lot!

[westerman]

I don't think that there is a program out there that will do what you want. In part this is because FastA headers are not standardized. Basically all you need for FastA is the '>'. So it would be hard to create a general purpose program that would combine FastA sequences together based on some regular yet arbitrary criteria. That said it would be trivial (for a programmer, at least) to create a one-shot specific program that would do the combining.

Now that I think of it, such an program would be a good one to give to a beginning programmer. Straight-forward yet not so trivial as to be boring.

[JackieBadger]

A little fiddly but you can use the text manipulation tool in Galaxy.

Convert your fastas to tabula formats....and use the cut column and paste files side by side to get your sequences in the same file.... you can then merge columns and convert back to fast

[westerman]

Jackie: I'll agree that conversion to tabular is a good first step -- and this can be done via the command line as well using the FastX tools -- but I don't see how a merger could be done automatically. Let's say after the conversion I have

Gene 1 Intron 1 <tab> GTACGCC....CTGATAGAG
Gene 1 Intron 2 <tab> GTCCAGGAC.....CTGAGTAAG
Gene 2 Intron 1 <tab> sequence
Gene 3 Intron 1 <tab> sequence
Gene 3 Intron 2 <tab> sequence
Gene 3 Intron 3 <tab> sequence
etc.

How to pull all of the Gene1, Gene2, Gene3, etc. sequences together without some programming? I'm not familiar enough with Galaxy to follow your "cut column and paste files" suggestion but if it is anything like the normal Unix 'cut' and 'paste' commands then I just don't see how to do the cut'n'paste automatically.

Perhaps if the tabular file becomes:

Gene<tab>1<tab>Intron<tab>1<tab>Sequence
Gene<tab>1<tab>Intron<tab>2<tab>Sequence
etc.

Then maybe cut'n'paste would work.