GO analysis of RNAseq Assembly

[ikim]

Hi all. I have a plant transcriptome assembled using Trinity that I have annotated with interpro and mapped GO ids to the individual transcripts such that I have a list such as :

comp1_c1_seq1 : 0034308
comp1_c2_seq2 : 0061077
comp1993_c1_seq1: 0055044
... etc.

I wish at this point to analyse and visualize the distribution of these ontologies. i.e. quantity by category/subcategory.
I know Blast2go has such functions but I'd need re-annotate through their pipeline. The only software I've found that performs this is Wego, but this is a web based and max sequence limit means I need to separate my assembly set into 8 subsets.
Can anyone recommend a tool that will take a contig/GO list and generate abundance distributions from it? Thanks very much!

[marnilush]

Hi Ikim,

i fear i do not have tool that does exactly what you asked for,
but since you are working with a plant transcriptome you might
be interested in our mercator pipeline
(http://mapman.gabipd.org/web/guest/app/mercator)
which was developed exactly for functional annotation of plant
transcriptome/proteome data.

It will compute functional annotations for each entry transcript
based on the MapMan ontology (that is widely used in the plant field).
One of the standard outputs of the pipeline in a chart showing exactly
the output you want - a distribution of your input sequences in
functional categories. You would have to run your raw sequences
through the pipeline again to obtain the results - however, this will
finish quite quickly (usually less than one day - depending on the size
of the input data set).

If your data set is too large to be submitted via the web interface,
please feel free to contact me directly ([email protected])
and i will submit the data manually.

If you need/want to have GO terms as well, we provide a MapMan
to GO translation table in our online repository (BIN2GO - available
via http://mapman.gabipd.org/web/guest/mapmanstore)

I hope this helps you,

best greetings,
Marc

[ikim]

Hi Marc,

Thanks very much for your advice. Your pipeline looks very impressive; it runs all of the searches which we use for annotation; i.e. much more indepth than say blast2go. Unfortunately I have close to 100 sets and rerunning their entire annotations for the purpose of capturing a summary image isn't practical at this point. I've just figured out that it's possible to generate .annot files that list contig-GO-EC-Description and import these into blast2go, therefore skipping their search and annotation step, and just using their graph plotting and visualization functions. Again thanks for your response!

Cheers,

Michael

[lzsph]

Hi Marc,

Thanks very much for your advice. Your pipeline looks very impressive; it runs all of the searches which we use for annotation; i.e. much more indepth than say blast2go. Unfortunately I have close to 100 sets and rerunning their entire annotations for the purpose of capturing a summary image isn't practical at this point. I've just figured out that it's possible to generate .annot files that list contig-GO-EC-Description and import these into blast2go, therefore skipping their search and annotation step, and just using their graph plotting and visualization functions. Again thanks for your response!

Cheers,

Michael

Hi Michael,

I also got our annotation file (not generated through Blast2GO pipeline) formatted into .annot format according to the Blast2GO's example. Besides the GO functional classification, can I directly run the "Annotation" --- "Enzyme Code and KEGG" --- "Load Pathway-Maps from KEGG (online)" to generate KEGG pathways?

Thank you.

Regards,
lzsph