On Sun, Feb 17, 2013 at 7:25 PM, Rachel North wrote:

I have been having some issues with Phenix. When I input the mtz file (generated from CCP4), an error message appears... "File format error: cctbx Error: invalid combination of values while extracting anomalous array from columns: F New, SIGF New, DANO New, SIGDANO New, hkl= -3,25,44."

This happens when I try to use the reflection file editor, xtriage and phenix refine. I have tested datasets collected from other crystals in the same condition and still have the same problem. I have also tried to process my data using different programs (XDSME and iMOSFLM) before going to CCP4 and then Phenix for refinement.

I have never had this problem before and I have run every program on CCP4 the same way that I usually do. Does anyone know how I fix this or where I am going wrong?

(I already replied to the message you sent to [email protected], but since it's looking like a systematic problem, here it is again for the benefit of anyone else who encounters it. If someone knows how and why these files are being generated, I'd love to know...) Several users have reported this problem recently - I suspect one of the scaling or conversion programs has changed and is now producing reflections with nonsensical values. Specifically, in the examples I've seen, F and SIGF are null but DANO and SIGDANO are valid numbers - not sure how this is possible, it could be a bug (or perhaps I'm misunderstanding the meaning). Anyway, we changed the code a week or so ago to discard these reflections instead of stopping. If you download the latest nightly build it should have the fix: https://www.phenix-online.org/download/nightly_builds.cgi -Nat