Reliability of short read annotations
Hi,
I would like to know how reliable annotations are for metagenome short reads generated by illumina sequencing ~70-100bp when reads are annotated for function (i.e. via MG-RAST, MEGAN etc)? For example if an alignment based on the short reads matches specific proteins should one be cautious to then say that those specific proteins belong to the organisms coming up in the alignments matches? Or should such functional assignments only be used to get a general sense about the function of the gene?
I would like to know how reliable annotations are for metagenome short reads generated by illumina sequencing ~70-100bp when reads are annotated for function (i.e. via MG-RAST, MEGAN etc)? For example if an alignment based on the short reads matches specific proteins should one be cautious to then say that those specific proteins belong to the organisms coming up in the alignments matches? Or should such functional assignments only be used to get a general sense about the function of the gene?
