Dear colleagues,
I am running a test of my local version of I-TASSER-5.1. I have submitted jobs for several proteins both on local package and on web-server. However, I have found out that the final models differ sufficiently (not as rotamers of side-chains). I have updated the ITLIB last week, and downloaded the full 89 GB of the libraries. Where can the problem come from? I am running the gnu-parallel calculations.
Best regards,
Dmitrii