***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***


SigFP = 0 is the historical way to indicate that the reflection is absent. Just create a column with all SigFP set to 1.0 (or to FP/10) would keep REFMAC and other programs happy. Just take care you don't run programs that will actually be using the fake SigFP data (SigFP = FP / 10 would perhaps be a decent choice if you need to do this).

Bart

PS to make the SIGFP column you can use SFTOOLS

sftools
read old.mtz
calc Q col SIGFP = col FP 10 /
write new.mtz
exit

(The Q sets the column label to Q which, if I remember well, is the label for a standard deviation)

Donnie Berkholz wrote:
***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***


Hi all,

I've got a 1.5A pdb solved about 20 years ago as well as the hkl file for it (with all sigfp's = 0), but I'm having trouble refining it. I want to refine it to calculate some phases to create a map.

I've imported it into an mtz and tried to run a round of refinement in Refmac, but it's giving me this error:

"No reflections in resolution bin ???????"

Somebody else ran into this a year ago [1], but their problem seemed to be with test-set generation. I generated the freeRflags twice, and both times hit the same error. After exporting the mtz to an hkl again with mtzdump, everything looked fine.

I tried narrowing down the resolution to just refine in 5-2A, but still got the same issue. I also tried including the freeR data, but that also didn't help.

Any ideas?

Thanks,
Donnie

1. http://www.dl.ac.uk/list-archive-public/ccp4bb/2005-01/msg00399.html

Reply via email to