Hi Phenixers, I have copied mtz file to a new file with a lower resolution limit using the "Reflection file editor” tool. I copied all arrays to the output file. When I attempted to continue refinement with the new mtz file, a message appears saying that there are no R-free flags in the mtz file. Yet, the new file contains the R-free-flags column but the values are inverted, 1 is changed to 0 and 0 is changed to 1. In the output option dialog the box “Preserve original flag values” was checked. How to fix this? Thanks, Mirek
Dear Mirek, it is a stubborn myth the Free-R flags need to be conserved. You can simply regenerate a new set of flags. This does not compromise the free F-value. This is based on what I would call Tickle's conjecture, even though with one of his latest emails on the ccp4bb, where he explained this in an extremely beautiful manner, it is not a conjecture anymore, rather a corollary. It is sad that even some well established crystallographers stick to said myth. Best regards, Tim cc ccp4bb for information Am 25.06.2019 04:09, schrieb Cygler, Miroslaw:
Hi Phenixers, I have copied mtz file to a new file with a lower resolution limit using the "Reflection file editor” tool. I copied all arrays to the output file. When I attempted to continue refinement with the new mtz file, a message appears saying that there are no R-free flags in the mtz file. Yet, the new file contains the R-free-flags column but the values are inverted, 1 is changed to 0 and 0 is changed to 1. In the output option dialog the box “Preserve original flag values” was checked. How to fix this? Thanks,
Mirek _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb Unsubscribe: [email protected]
-- -- Tim Gruene Head of the Centre for X-ray Structure Analysis Faculty of Chemistry University of Vienna Phone: +43-1-4277-70202 GPG Key ID = A46BEE1A
Hi Mirek, based on your description this sounds like a bug or undesired feature at least! Could you please send me the files so that I can repeat your experience myself and see what's happening? As a work-around you can re-generate flags as Tim suggested but chances are someone annoying like myself asks to prove no memory left from previous set and that is very hard to prove, and if not proven there will be a doubt! Although indeed, re-generating and running some refinement should be just fine, I agree! Good luck! Pavel On 6/24/19 21:09, Cygler, Miroslaw wrote:
Hi Phenixers, I have copied mtz file to a new file with a lower resolution limit using the "Reflection file editor” tool. I copied all arrays to the output file. When I attempted to continue refinement with the new mtz file, a message appears saying that there are no R-free flags in the mtz file. Yet, the new file contains the R-free-flags column but the values are inverted, 1 is changed to 0 and 0 is changed to 1. In the output option dialog the box “Preserve original flag values” was checked. How to fix this? Thanks, Mirek
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb Unsubscribe: [email protected]
Hi Pavel, if you would ask that question, you did not read Ian's explanation properly. His email reads like a proof, and anyone doubting the validity make claim without understanding. On the contrary: what evidence do you provide that there was a memory effect? In case you trust experimental evidence better than mathematical rationale, you can repeat the experiments described under "Validation I: How "free" is Rcomplete" from www.pnas.org/cgi/doi/10.1073/pnas.1502136112 with any data set of your choice. Our paper also describes experiments to provide evidence that Rcomplete is equivalent to Rfree (or rather better in case of small data sets). I hope that if ever you make such a claim as reviewer, you can provide a literature reference with evidence of your claim, because reviewers that make claims without references are very time consuming (and often just proove their incompetence). Best, Tim On Tuesday, June 25, 2019 9:22:37 AM CEST Pavel Afonine wrote:
Hi Mirek,
based on your description this sounds like a bug or undesired feature at least! Could you please send me the files so that I can repeat your experience myself and see what's happening?
As a work-around you can re-generate flags as Tim suggested but chances are someone annoying like myself asks to prove no memory left from previous set and that is very hard to prove, and if not proven there will be a doubt! Although indeed, re-generating and running some refinement should be just fine, I agree!
Good luck! Pavel
On 6/24/19 21:09, Cygler, Miroslaw wrote:
Hi Phenixers, I have copied mtz file to a new file with a lower resolution limit using the "Reflection file editor” tool. I copied all arrays to the output file. When I attempted to continue refinement with the new mtz file, a message appears saying that there are no R-free flags in the mtz file. Yet, the new file contains the R-free-flags column but the values are inverted, 1 is changed to 0 and 0 is changed to 1. In the output option dialog the box “Preserve original flag values” was checked. How to fix this? Thanks, Mirek
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb Unsubscribe: [email protected]
-- -- Tim Gruene Head of the Centre for X-ray Structure Analysis Faculty of Chemistry University of Vienna Phone: +43-1-4277-70202 GPG Key ID = A46BEE1A
participants (4)
-
Cygler, Miroslaw -
Pavel Afonine -
Tim Gruene -
Tim Grüne