CDS sequence from mRNA and protein sequence

Question

[Matteo Schiavinato]

Hello guys,

Here's for the (probably) dumb question of the day, but I can't seem to find fitting solutions in a reasonable amount of time, so I might be missing something here.

I have an mRNA multifasta and a protein multifasta (IDs are correspondent between the two). What I'm after, is to get the CDS sequences from the mRNA file, using the proteins as model for that. Or any other way to get the CDS sequences from those mRNAs. I do not have genome coordinates or any type of GFF file.

Any easy way that you might know?

[Benn]

You can try blastx (with the protein fasta as db), I assume your mRNA sequences are without introns?

[Matteo Schiavinato]

Yes they are intronless!

[Sej Modha]

BLASTX should do the trick but not sure if you would be able to extract the exact coding sequences using it, however, you can certainly use the coordinates.

Answer 1

[lieven.sterck]

Simply run some software on it get the longest ORF (plenty of tools around for that) from the mRNA, should (in theory) end up to the proteins. Additionally just double-check them against your protein set (eg calculate MD5 key and compare both).

potential software to use:

• some custom python script How to extract the longest orf?
• EMBOSS getORF (not sure it it gets the longest one by default though)
• NCBI orfFinder
• TransDecoder
• FindLongestOrf.pl
• max_orf

and many other around I guess

[Matteo Schiavinato]

I am interested in which software would be the way to go!

[lieven.sterck]

OK, I'll add some to my 'answer'

My general advise would be to go for one that does nothing more than to look for the longest orf, so preferentially none that use coding potential or such but simply (and bluntly) gets the longest one from a potential start to a stop (by limiting to the forward strand even)