Aligning RNA-Seq reads in peptide space

[Blahah404]

I'm planning to do some cross-species comparative transcriptomics, with non-model plant (grass) species. I'll be aligning the reads to a closely related genome (same tribe), evolutionarily equidistant from the experimental species. As far as I can tell, the tools which are commonly used for aligning RNA-Seq reads do so in nucleotide space. Are there any tools which can work in peptide space? Or can anyone suggest a strategy using existing tools which would work for cross-species alignment?

I'm not afraid to develop my own tools, but I don't want to do it if the tools are already there!

Many thanks

[mbblack]

I have not had cause to use it in a very long time, but t-coffee is one sequence aligner that will align a DNA sequence given the constrant of underlying known protein coding sequences.

I have no idea if anyone has adapted or applied it to batch processing a next-gen data set though.

http://www.tcoffee.org

[areyes]

maybe tblastx could help as well

[mbblack]

Again, I have no experience actually using these, but in my EndNote library a see

transAlign - published in BMC Bioinformatics. 2005; 6: 156. This is a PERL script tool.
MACSE - ("Multiple Alignment of Coding SEquences Accounting for Frameshifts and Stop Codons") published in 2011, PLoS ONE 6(9).

I just don't know of anything that has been specifically applied to a batch alignment of a next-gen readset.

[Blahah404]

@mbblack thanks very much - tcoffee looks interesting. From what I can make out, it first indexes all the pairwise alignments between each pair of sequences. I think when comparing a short read to a whole genome this would take forever and possibly makes that approach unsuitable. Nevertheless I'll ask on their google group and see if anyone has adapted it.

@areyes yeah, after I made the post I realised I should start by looking at BLAST and BLAT. I've been playing with BLAT using 454 reads, which seems to be working OK. I'll compare with tblastx too.

In general I think tools for local alignment will be more likely to be applicable than those for multiple alignments.

[mbblack]

If you find something that you could adapt, be sure to search a bit and see if it has been optimized further by someone other than the first or maintaining author. I cannot find the link, but there is an MPI parallel version of t-coffee around, for example.

[Blahah404]

The MACSE paper is really quite useful, thanks. Time to start testing tools!

[liruiradiant]

Hi, Mbblack! I'm a new guy in bioinformation field. I see that you mentioned Endnote. Is there an equivalent of Endnote in Ubuntus please? Thanks

[Blahah404]

@liruiradiant

1. nobody mentioned Endnote, but an excellent open-source alternative is Zotero (http://www.zotero.org).
2. you should start your own thread when you have a new question, otherwise very few people are likely to read it.

[mbblack]

Hi, Mbblack! I'm a new guy in bioinformation field. I see that you mentioned Endnote. Is there an equivalent of Endnote in Ubuntus please? Thanks

I referenced some alignment tools I have in my Endnote reference library.

I do not know about any reference managers for Linux as the only things I use in Linux/UNIX are command line tools (I don't use any desktop software at all on my Linux boxes), sorry.

[liruiradiant]

Thank you for your reply. Maybe I'll do the same with you: Have one lab computer running Linux and my laptop Windows for endnote. Sometimes it's just not so convenient. But it's a problem without solution I think. Thanks anyway!

I referenced some alignment tools I have in my Endnote reference library.

I do not know about any reference managers for Linux as the only things I use in Linux/UNIX are command line tools (I don't use any desktop software at all on my Linux boxes), sorry.

[sphil]

Zotero is a good tool, also JabRef is available for both platforms (also for Mac)... simple but it will do the job as a reference mangager...Jabref also exports BibTex...