Talk:Task 6 - Protein structure prediction from evolutionary sequence variation

From Bioinformatikpedia
Revision as of 23:10, 1 September 2013 by Betza (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Hi all, my problem is that for both pfam domains [1] [2] of our protein HFE_HUMAN, EVcouplings and EVfold stop at the mapping step without providing further details about the problem.

EVcouplings input for the MHC domain was:

SHSLHYLFMGASEQDLGLSLFEALGYVDDQLFVFYDHESRRVEPRTPWVSSRIsSQMWLQLSQSLKGWDHMFTVDFWTIMENHNHSKESHTLQVILGCEMQEDNSTEGYWKYGYDGQDHLEFCPDTLDWRAAEPRAWPTKLEWERHKIRARqNRAYLERDCPAQLQQLLELGRG
known structure: 1A6Z

Also, I don't get emails about the job statuses anymore, when i start a job. For Ras on the other hand, everything worked fine.

Another question

What value can the CN score maximally assume?

From my results i would guess that it is below 10 in most cases. But since the freecontact paper is not released yet, I don't know how it is calculated.

Best, Alex