How to deal with high percentage of non-specific matching during read mapping on Trinity assembly?

Question

[seta]

Hi everybody,

Regarding Trinity tool for transcriptome assembly, I noticed that it has high percentage of non-specific matches during mapping reads back to assembled transcriptome. It may almost normal as Trinity generates gene isoforms, too. But, I think it will have an adverse effect on downstream analysis such as gene expression analysis, am I right?. Could you please let me know how you deal with this issue? I also tried reduce redundancy using cd-hit-est tool, but it had not a significant effect on the issue. Looking forward to hearing your solution and experiences.Thanks

Answer 1

[h.mon]

Regarding gene expression, Trinity includes Perl scripts to analyze differential gene expression, and the scripts include workflows for "transcript" and/or "gene" levels. The "gene" level analysis is what you want, as it aggregates counts from all isoforms Trinity identified into one "gene".

[seta]

Thanks. Assuming we want to pool Trinity assembly with another assembly, is there any way to reduce the redundancy before merging assemblies in order to make a non-redundant transcriptome assembly? I used cd-hit-est tool to this end, but I don't know why it was not perfect in this regard, non-specific matching still remained.