How to fully use the list of DEGs in non model species?

[jack Ip]

Hi ALL,

I am a beginner of RNAseq study. I am working on the transcriptome of non model species, mussel and snail. In my first part of study, I have investigated the transcriptome of my target species. And then, I have done exposure experiments and RNAseq to determine the toxicity mechanism.

Now I have performed DESeq2 and EBSeq, I have a list of DEGs. As the molecular information of my target species is limited, most of the assembled unigene/transcript don't have any functional annotation, and BLAST hit to NCBI nr. Say only 20% DEGs have their BLAST hit to public databases.

I would like to ask how can I fully use the DGE data to reveal the toxic effect and mechanism? How can I deal with the 80% DEGs without annotation?

I also performed GAGE using the whole transcriptome at background. again the unannotated may cause the few or no enriched pathway/GO returned from GAGE ><"".

Please kindly give me some suggestions of data analysis and how to utilize the result data.

Many thanks,
Jack

[Oyster_lab]

Hi Jack,

I'm having the same issue and would like to hear some suggestions on how to move further. I have ~40% of Blast hits, many of them with predicted/hypothetical proteins.

How did you perform GAGE? I'm trying to run it using my DGE results from EdgeR. How have you created the "kegg.gs" file?

Thanks for your help!
Cheers,
R