translate trinity.fasta or use transdecoder.fasta.pep as protein database?

[Cheryl Ames]

Hello Trinityrnaseq-users:

I work on a non-model organism, and want to use my transcriptome (trinity.fasta) file to make a protein database for a downstream proteomics project on the same species.

I am exploring these two options:

Option 1: query my protein sequences against the trinity.transdecoder.fasta.pep

Option 2: translate the entire trinity.fasta file into amino acids using all 6 reading frames (3 forward; 3 reverse) in Geneious.

Logically which option would work best as to generate a "protein database" for my non-model organism?

With option 2, I would need to reformat the forward and reverse strands in order to distinguish them in a concatenated database.

With option 1, it is concerning that the  transdecoder.pep file is much smaller than the translated and concatenated trinity.fasta database will be.

Does anyone have a better option?

Suggestions are appreciated.

Best,

Sheriru

[Brian Haas]

Hi Sheriru,

For proteomics studies, folks usually do the 6-frame translations and search that.  This way, you'll not miss anything that was otherwise excluded by the trandecoder search

best,

~brian

--
You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-u...@googlegroups.com.
To post to this group, send email to trinityrn...@googlegroups.com.
Visit this group at http://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

[Cheryl Ames]

Dear Brian,

Thank you.

In order to search the six-frame translation effectively, can you recommend at good way of:

1. Changing the format of the reverse translation so that entries there can be distinguished from those from forward translation?

--> Just adding an F or R at the end of the contig?

2. Lining up the format so that the search engine can use one parsing rule for all entries?

--> Just concatenating all the files into one?

Any suggestions would be great.

Best,

Sheriru

[Brian Haas]

Hi Cheryl,

Attached is a script that will do the 6-frame translation of a fasta file.  You can drop it into TRINITY/util/misc  and it should work.

It generates output like so, providing F${frame}_trans as a suffix to the accession:

>TR83|c0_g1_i1.F1_trans TR83|c0_g1_i1 len=140 path=[235:0-139] [-1, 235, -2]
RRHNQGTRLRGTDASRSGRTAAERKRRRRRCQRQPRAAAAPRGKGR
>TR83|c0_g1_i1.F2_trans TR83|c0_g1_i1 len=140 path=[235:0-139] [-1, 235, -2]
ADIIRAPG*EEPTRHAAAERLRRGSGGGGGASASPELPRRREARGG
>TR83|c0_g1_i1.F3_trans TR83|c0_g1_i1 len=140 path=[235:0-139] [-1, 235, -2]
PT*SGHQVKRNRRVTQRPNGCGEEAAAAAVPAPAQSCRGAERQGAE
>TR83|c0_g1_i1.F4_trans TR83|c0_g1_i1 len=140 path=[235:0-139] [-1, 235, -2]
LRPLPLGAAAALGWRWHRRRRRFLSAAVRPLRDASVPLNLVP*LCR
>TR83|c0_g1_i1.F5_trans TR83|c0_g1_i1 len=140 path=[235:0-139] [-1, 235, -2]
SAPCLSAPRQLWAGAGTAAAAASSPQPFGRCVTRRFLLTWCPDYVG
>TR83|c0_g1_i1.F6_trans TR83|c0_g1_i1 len=140 path=[235:0-139] [-1, 235, -2]
PPLASRRRGSSGLALAPPPPPLPLRSRSAAA*RVGSS*PGALIMSA

I hope this helps,

~brian

--