Does the synonyms file in MEGAN work for the SEED/KEGG Analyzers?

[nrnoyes]

I am trying to run my Silva BLAST results through the SEED and KEGG Analyzers in MEGAN. I understand that the blast file needs to have RefSeq ID's in order to work with the analyzers. It seems there is a "synonyms file", provided by MEGAN, to use with Silva results in order to map correctly; however, it is unclear if this is just for the taxonomic analysis, or also works for the SEED/KEGG analysis...

I've tried using the synonyms file, and I do get KEGG/SEED results, but they consist purely of "no hits" and "unassigned", so something isn't quite right...

Any help would be much appreciated! Thanks!

[RickBioinf]

When I was working with this I found out that the "synonyms file" is not always quite up-to-date with the Silva database. What you could do is update it yourself or ask the MEGAN guy to do this for you.

[nrnoyes]

Thanks RickBioInf! If I do update the synonyms file myself, would it work with the SEED/KEGG Analyzers, or do these tools require a blast file that has the RefSeq ID's directly in it? It seems from the MEGAN manual that perhaps the synonyms file only works with the taxonomic classification, NOT with the functional classification tools (i.e. SEED/KEG)...

[cliffbeall]

Am I missing something? I think you might need different data to get SEED/KEGG info.

The SILVA I am aware of is an rRNA database. The SEED/KEGG functions in MEGAN are for analyzing mapping of shotgun genomic reads to protein databases like Genbank nr.

[nrnoyes]

cliffbeall -- You are correct. Sorry, very new to this. I believe NCBI-nr must be used for both KEGG and SEED, in order to ensure that MEGAN can map the RefSeq ids to the functional annotation. Thanks!

[nrnoyes]

Hi again -- another related question. Have you tried running the taxonomic classification on Silva output without the synonyms file? It gives output, but the output is different than when I use the file (using exact same LCA parameters)...