Constructing a gene-to-trans map for a non-trinity assembly to use in RSEM

[John Henske]

I have a reference transcriptome that was not assembled with trinity that I would like to use for RSEM analysis. I tried to create a gene_trans_map file, but could not make one that worked. What is the proper format for this input file? 

All that is mentioned about the format is that it should be a file containing 'gene(tab)transcript' identifiers per line.

If I have the following 5 transcript IDs (where LocusX indicates the gene)

Locus1v1rpkm45399.37

Locus2v1rpkm41529.00

Locus2v2rpkm3172.69_PRE

Locus3v1rpkm32852.55

Locus4v1rpkm28088.74

Should the file look like this?

gene          transcript

Locus1      Locus1v1rpkm45399.37

Locus2      Locus2v1rpkm41529.00

Locus2      Locus2v2rpkm3172.69_PRE

Locus3      Locus3v1rpkm32852.55

Locus4      Locus4v1rpkm28088.74

Thank you,

John

[Brian Haas]

Yes, that should work.

good luck!

~b

--
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.

[naveen sharma]

hi john. i have the similar problem. I just want to know that i have transcript fasta file which is not assembled by trinity, can i use that file for RSEM analysis?? I tried using it without giving the trinity_mode and it is running. Please help me in this regard

[Brian Haas]

I'd suggest looking at CORSET:

http://www.genomebiology.com/2014/15/7/410

~b

--