Hi,
I have few questions regarding secondary structures reported by LOMETS, MUSTER and I-TASSER servers.
First, I compared MUSTER, LOMETS and I-TASSER results and I've noticed that they reported different secondary structures. More precisely, LOMETS and I-TASSER have the same structure and MUSTER's differ a little. How this influence the final model?
Jobs:
MUSTER results: http://zhanglab.ccmb.med.umich.edu/MUSTER/output/S28076/
LOMETS results: http://zhanglab.ccmb.med.umich.edu/LOMETS/output/S64063/
I-TASSER results: http://zhanglab.ccmb.med.umich.edu/I-TASSER/output/S208138/
Second, I saw that MUSTER, LOMETS and I-TASSER report PSI-PRED secondary structure, so I compared these results with PSI-PRED server and I noticed that they differ. Is this due to different versions of PRIPRED and how it influences results?
PSIPRED results: http://bioinf.cs.ucl.ac.uk/psipred/result/101e5078-9dad-11e4-9475-00163e...
Finally and most important, I compared LOMETS and I-TASSER top models and I've noticed that LOMETS better preserved secondary structures. Why is that and what can I do in order to increase content of secondary structures in I-TASSER result with preserving the best template?
Thank you very much for your help