On Mon, 2010-11-29 at 14:30 +0000, Jyotica Batra wrote: > is there a way I can switch to phenix.refine by retaining the same > R-frees, I have from refmac
If you actually mean the test set, then you don't need to do anything other than use the same mtz-file. Keep in mind though that in phenix by default (unless it changed recently) the intensities will be converted to amplitudes internally using simple square root (i.e. no French&Wilson) and negative intensities discarded. Hence the safe way to minimize changes would be to make sure that your mtz-file only contains Fo's and not intensities (there might be also some phenix keyword to rectify this, which I am sure you will be promptly informed about). But if you really mean getting the same R-free, I am afraid it's almost impossible and rather pointless. There are many things the two programs do differently (measurement outlier rejection, bulk solvent parameters refinement, maximum likelihood target construction, default restraint targets, and even the target minimization procedure itself, just to name a few). No intent to start another round of the quarterly "who cares about R-values" discussion. Carthago delenda est. Cheers, Ed. -- "I'd jump in myself, if I weren't so good at whistling." Julian, King of Lemurs
