Hello Diego-- > > I'm currently working on a rigid body protocol to calculate > ensembles of structures for my three-domain protein to optimize the > fit with my SAXS data. However, I've encountered an issue with the > attached script when attempting to modify/optimize the weights of > each ensemble component, resulting in an error message. Could you > please help me with the script?
The error message would be appreciated- the problem is not immediately evident to me. > > Regarding ensemble modeling with SAXS data, I have some questions: > > 1. When determining the ensemble size, is it preferable to use an > even number, or is the choice of parity irrelevant? This depends on the problem. If you suspect there are primarily three states then Ne=3 is a good choice. If you want to sample a large number of Ne values, a sparse sampling protocol might be a good idea. For computational speed, some thought might be necessary- like how cores are laid out within a CPU or in a multi-processors machine. Presumably, you are using the -num_threads N options to parallelize over the ensemble. > 2. I've noticed discrepancies in the chi2 values between the xray > and xray-c potentials. Could you clarify the reason for this > difference, and which value corresponds to the one calculated by > calcSAXS or crysol? Hopefully, the chi^2 values are close- as the xray values are fixed up whenever xray-c is recalculated. The xray-c values should be more accurate, and should correspond to those reported by calcSAXS more closely. > 3. Within the script, would it be more advantageous to employ > repel and torsionDB potentials instead of VDW and RAMA? torsionDB is better than RAMA, and configuration of RepelPot is simpler than using VDW. best regards-- Charles ######################################################################## To unsubscribe from the XPLOR-NIH list, click the following link: http://list.nih.gov/cgi-bin/wa.exe?SUBED1=XPLOR-NIH&A=1
