Hello all,

We are currently working with a dataset for a trimeric protein complex at 3A that is giving us some issues.  The space group seems to be P21 ( 59.5, 61.6, 99.6 , 90, 98, 90) and the Mathews coefficient (2.6) suggests 2 copies of this 35kDa complex in the ASU.  We are using a homolog as the search model that is 80% identical, and Phaser is able to find 2 molecules in P21 when searching in the same point group and gives good TFZ scores (TFZ = 9 or 10).  However, these solutions do not refine well following Phaser (Rfree ~.50), and after removing parts of the model that look incorrect the Rfree still stays between .46 and .50.  We ran the data through Xtraige and saw no signs of twinning based on the L-test, and the Patterson map has one small off-origin peak that is only 5%, indicating that there is no significant pseudotranslation.  We have also tried finding MR solutions in P1, but haven’t had much luck.  Any ideas on where to go from here would be greatly appreciated.