Greetings again. First, I want to thank everyone for all the help from my previous post. Here is an update on my progress and I would appreciate further advice.
I have implemented (nearly) all of Dr. Clore's very useful suggestions. Specifically, I removed all stereospecific assignments from my NOE list. I continue to calibrate my NOE intensities using the CALIBA command in CYANA, but I lengthened the reference distance from 4.0 to 4.8 Ang. Now, my restraints are quite loose. I also add the extra 0.5 Ang for each methyl involved in a distance restraint. I get much fewer NOE, CDIH and VDW violations. Though, I still don't have it down to zero. As you can see from the attached "sa.stats" file, the RMSDs for all the violations are very low. Do people think that this is an acceptable number of violations? At the moment my biggest concern is with the large RMSD for the final ensemble of fitted structures. When I include only the ordered residues (excluding unstructured N and C termini), I still only have about a 2.5 Ang RMSD for backbone atoms. I am surprised it is so large. For comparison, and to remind you from my previous post, I originally assigned all these NOEs using NOEASSIGN (formerly known as CANDID) in CYANA. The final ensemble of structures from CYANA were very tight, around 0.5 Ang RMSD for backbone atoms. Granted, I suspect they were TOO tight, due to overly restrictive NOE distance restraints. However, I didn't expect for the structures to become this loose in Xplor-NIH after loosening the NOEs, especially because I added NH RDC restraints, some HBDA (h-bond) restraints and used the RAMA potential term. Additionally, I am concerned because the secondary structure elements are not "well-formed", i.e. the beta-strands and the alpha-helices are not fully formed and regular. The final CYANA output structures had better formed backbone secondary structure. I appreciate the importance of the improved accuracy afforded to NMR structures through the use of RDC restraints and the RAMA potential, but I need to better understand these differences I am seeing between CYANA and Xplor-NIH. I really don't get why my Xplor-NIH structures are poorly defined, especially since there are so many published structures determined by Xplor-NIH with similar numbers of restraints that are much tighter and look much better. Again, I suspect I am not implementing something correctly. On a slightly different topic, this is my first time using RDCs in structure determination. I measured HN, HNCO, NCO, NCA and CACO RDCs in Pf1 bacteriophage. The NCA and CACO data is clearly too low S:N, but the HN, HNCO and NCO look pretty good. I implemented the HN RDCs in Xplor-NIH and I immediately got good agreement with the NOEs; it was very encouraging. However, the HNCO and NCO RDCs do not agree at all. I think I scaled them correctly (approximately 3-fold for HNCO and 8-fold for NCO relative to NH). Currently, I am trying to recollect the HNCO and NCO RDCs with much higher S:N and slightly better 15N axis resolution. We didn't get good results with the PEG(C12E5):Alcohol mixtures (massive broadening), but I've ordered some Cetylpyridium to try that method. I suspect this domain binds lipid, so I am avoiding bicelles at the moment. I am told that if I can get just NH RDCs in a different alignment medium, this should help a lot. Again, thanks for all the help. This sort of discussion is very useful and I am learning a lot. Regards, Mike - Michael E Hodsdon, MD, PhD Assistant Professor of Laboratory Medicine and Pharmacology Yale University School of Medicine email: michael.hodsdon at yale.edu http://info.med.yale.edu/labmed/faculty/hodsdonm.html 203.737.2674 (office) 203.737.2684 (lab) 203.915.0780 (mobile) 203.688.8704 (fax) Administrative Assistant: Vilma Moreno phone: 203.737.5934; email: vilma.moreno at yale.edu U.S. Postal Address: P.O. Box 208035, Laboratory Medicine Yale University School of Medicine 333 Cedar Street New Haven, CT 06520-8035 Shipping Address: CB462b, Laboratory Medicine 200 South Frontage Road New Haven, CT 06520 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://dcb.cit.nih.gov/pipermail/xplor-nih/attachments/20070324/15950990/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: sa.stats Type: application/octet-stream Size: 25324 bytes Desc: not available Url : http://dcb.cit.nih.gov/pipermail/xplor-nih/attachments/20070324/15950990/attachment-0001.obj
