Discrepancy interpretation in hits number according to certain databases

[Oscar Aguinaga]

Hi all

I am working with 20 metagenomes and I have found a significant difference in one of my metagenomes when using certain databases. For example, if I use Subsystems database, I obtain ~ 2000 - 3000 hits per metagenome, everything ok there. However if I use the KO database, one of the metagenomes gets less than 200 hits.

Even more intriguing, when analyzing taxonomy, if I use Greengenes, no problems, but with Silva, the same metagenome gets less than 200 hits. I am working with many replicates, so that particular metagenome has 4 other replicates and they show no discrepancies. How should I interpret this? Any ideas?

Regards,

Oscar