Finding orthologous gene within species

Question

[Prasad]

Dear all,

I have checked many post in biostar and other blogs to find orthologous and paralogous gene identification across/within the species. Most of them are suggesting for Best Reciprocal Blast Method(BBRM). But i have 2 leaf and 2 rhizome transcript samples. I have to find out orthologous and paralogous genes between these set. Shall i use BRBM method or are there any other better way to do it.

Please suggest any tools.

Thanks all

[Whoknows]

Hi, You just download NCBI blast toolkit then make blast database via makeblastdb and finally search your sequences within target set by blastn (DNA) or blastp (for protein) ,

but i think NCBI/Blast can handle your job as well.

[Prasad]

thanks guys for the suggestions

[Denise CS]

There are resources with orthologues and paralogues already calculated that may be useful to you, e.g. orthologues to the A. thaliana PAD4 gene. Ensembl Plants has 44 species.

Answer 1

[Prasad]

Thanks for the suggestion.

If i am not mistaken, you are suggesting BRBM (Best Reciprocal Blast method) to use. If so i will find orthologous genes between samples. Could you please suggest any way to find paralogous genes as well

[Bioinformatics_NewComer]

For paralogous, you will blast your sequences against themselves. The first hit other then - gene itself and its isoform [if present] would be the best hit. That will be its paralogous hit.

Answer 2

[Whoknows]

Hi, You just download NCBI blast toolkit then make blast database via makeblastdb and finally search your sequences within target set by blastn (DNA) or blastp (for protein), but I think NCBI/Blast can handle your job as well.

[qiyunzhu]

Agree with this comment, and my two cents are that: The problem Prasad faces is that two leaf and two rhizome are not treated as different organisms in the default NCBI database, in which case, he has to re-define taxonomy. The straightforward way one can think about is to make BLAST databases by their own. Simply prepare four "genome" files and make them a BLAST database will do the job.

Answer 3

[konstantinkul]

Hi! Check OrthoMCL. It is the best way to find orthologous genes.

Answer 4

[Whoknows]

As I told you for those works you can use NCBI Blast, normally it shows different blast hits among various species.

Answer 5

[Biogeek]

OrthoMCL / evidential gene.

Also conditional reciprocal best blast / RBH are good methods for determining pretty confident orthologues. Paralogues may be harder to detect I guess.

[Biogeek]

Mind you I think RBH/CRBB is the easiest way forward.

https://github.com/peterjc/galaxy_blast/tree/master/tools/blast_rbh -Use this python script.

It needs NCBI+ to be installed. Set NCBI+ in your job PATHs, voila.

Answer 6

[Whoknows]

Hi

what we do in plant studies is using COG/KOG analysis or cluster of orthologous groups of proteins.

You can upload your protein sequences into WebMGA and it gives you ~25 cluster of protein groups across different species.

Hope it helps you.

[Biogeek]

I'd be worried about the database date of that second link - WebmMGA - 2011, bit outdated?

[Whoknows]

Yes you are right but WebMGA is easy to use, but for more updated version you could use EggNOG database, which is recently updated.