KO/EC Assignment still a problem

[Caroline Fortunato]

Hello,

I submitted 13 metagenomes back in January that were in the batch that had to be re-run due to the incorrect KO/EC assignmen. Now I am looking at the newly re-run data and the percent of hits to the KO database is still extremely low. The percent of genes hitting COG and pfam runs between 50-70%, were as KO is 1-3%. For previously submitted metagenomes these percentages were nearly equal, so it seems very odd that for this batch of metagenomes the percentage of KO assignments are so low. 

If anyone could lend insight into why this might be that would be extremely helpful. Thank you,

-Caroline

[Manoj Pillay]

Hello Caroline,

                  There was an issue with our synchronization process. Upon reloading corrected annotation data, our production systems failed to get an update on precomputed statistics. I was able to force an update to enable the Internet facing databases in acquiring the latest statistics. Please note that all this while the underlying data was accurate but just not the precomputed stats such as counts and percentages. We apologize for this bug and thank you for bringing this to our attention - your report helped us correct the problem for over 300 datasets that were reloaded alongside yours. Please take another look and let us know if the recomputed statistics are in agreement with your expectations.

Thanks,
Manoj