problems when adding Free_R_flag
Dear experts I use *Reflection file editor* in phenix GUI to generate Free_R_flag in my mtz file. The original input .mtz file contains arrays of FP SIGFP DANO SIGDANO and ISYM. I just simply click *COPY ALL ARRAYS* and tag on *Generation R free flags if not already present*, then run the software. The problem is that in the output .mtz from phenix I found the value of F and DANO becomes more or less same for each reflection. This is unreasonable for a reflection file. Phenix changed something on the columns of my .mtz file. How? Why? What should I do to keep my .mtz in correct and unchanged form? Thank you very much for your advice. The columns in the original input .mtz file look like H K L FP SIGFP DANO SIGDANO ISYM 0 0 12 319.30 2.64 2.23 5.27 0.00 0 0 16 424.00 3.50 -8.38 7.00 0.00 0 0 20 177.60 1.48 1.77 2.96 0.00 0 0 24 155.10 1.30 3.97 2.61 0.00 0 0 28 195.80 1.64 1.13 3.28 0.00 The output from phenix looks like (FP and DANO become similar in value for each reflection) H K L FP SIGFP DANO SIGDANO ISYM FreeRflag 0 0 12 320.41 3.73 318.19 3.73 0.00 0.00 0 0 16 419.81 4.95 428.19 4.95 0.00 0.00 0 0 20 178.48 2.09 176.72 2.09 0.00 0.00 0 0 24 157.09 1.85 153.11 1.85 0.00 0.00 0 0 28 196.37 2.32 195.23 2.32 0.00 0.00
On Tue, Feb 14, 2012 at 11:56 AM, G Y
The problem is that in the output .mtz from phenix I found the value of F and DANO becomes more or less same for each reflection. This is unreasonable for a reflection file. Phenix changed something on the columns of my .mtz file. How? Why? What should I do to keep my .mtz in correct and unchanged form? Thank you very much for your advice.
This is a bug that we fixed very recently; please update to the latest nightly build. (The problem, FYI, is that we used to convert F,SIGF,DANO,SIGDANO,ISYM to F+,SIGF+,F-,SIGF- internally, which is also what would be output - unfortunately, the reflection file editor was keeping the old labels around. You could fix the ambiguity in the version you're currently using by changing the output labels appropriately, but I assume you'd rather keep the original data types. [Although we still think it's better to use the original Friedel mates; Phenix never uses the anomalous difference array by itself.]) -Nat
participants (2)
-
G Y
-
Nathaniel Echols