Peter,
Depending on the method implemented in one software or another - I wouldn't be surprised if some of them fall into local minima during gradient optimization for best fit. Depending on how you define the region of interest, there may also be ambiguity with respect to pair definition - this may not be relevant if you are looking at exactly same regions of exactly the same sequence. I recommend LSQMAN from the Uppsala software package, as final verification. Regarding plots of RMSD versus number of atoms - LSQMAN can be scripted (externally, via e.g. Perl) to do that. I wouldn't be surprised if Coot can also be scripted to do the exact same thing via Python. Artem "Nothing is built on stone; all is built on sand, but we must build as if the sand were stone" Jorge Luis Borges _____ From: CCP4 bulletin board [mailto:[email protected]] On Behalf Of peter hudson Sent: Sunday, July 05, 2009 2:33 AM To: [email protected] Subject: [ccp4bb] RMSD value is different from O and superpose Hello all I superposed the two different chain, which are two different subunits of the same structure which are different in conformation with two differen t programmes. 1) First is with O, with LSQ command, it gives the RMSD values for different chain is 1.5A for 108 ca atoms. 2) Second programme i used is Superpose from CCP4i inbuilt, which gives the RMSD values of 2.4A for almost same ca atoms. Why this two programmes output RMSD are very different. In my opinion the superposition by O seems much better, after looking at the structure. Apart from the difference in the RMSD, I would like to plot the RMSD values against the no. of ca atoms superposed. That i only know to do through superpose. Is there is any way to plot the RMSD values i get from O against the no. of ca atoms. I would appreciate the help. Thanks in advance. peter
