| Source | Coefficient | Acentrics | Centrics |
|---|---|---|---|
| Literature | FM | 2mFo - DFc | mFo |
| FD | 2(mFo - DFc) | mFo - DFc | |
| FC | DFc | DFc | |
| BUSTER | FM | 2mFo - DFc | 2mFo - DFc |
| FD | 2(mFo - DFc) | 2(mFo - DFc) | |
| FC | DFc | DFc | |
| REFMAC | FM | 2mFo - DFc | 2mFo - DFc |
| FD | mFo - DFc | mFo - DFc | |
| FC | DFc | DFc | |
| SIGMAA | FM | 2mFo - DFc | mFo |
| FD | 2(mFo - DFc) | mFo - DFc | |
| FC | DFc | DFc |
where the 'coefficient' is:
FM for the '2Fo-Fc' map, FD for the 'Fo-Fc' map and FC for the 'Fc' map.
Differences from the literature values are indicated in red. Note that for Refmac, if the purpose is only to compute the map, halving the acentric FD coefficient has the same effect as doubling the centric one (so Buster & Refmac are consistent in this respect). However if the statistical analysis program EDSTATS is to be used it is essential to correct this inconsistency, otherwise incorrect map coefficients will be computed internally by EDSTATS.
The consistency checks performed by this program are (A=acentric, C=centric, else both):
| A1 | FC = 2mFo - FM |
| A2 | FC >= 2mFo - FM |
| A3 | FD = FM - mFo |
| A4 | FD = 2(FM - mFo) |
| C5 | FC = 2mFo - FM |
| C6 | FM = mFo |
| C7 | FD = mFo-FC |
| C8 | FD = 2(mFo-FC) |
| C9 | FD <= mFo-FC |
| 10 | FM = FC (unobserved only) |
| 11 | FD = 0 (unobserved only) |
The program makes a first pass through the input MTZ file to perform the above checks and work out what needs to be fixed. Then if required, a second pass makes the necessary corrections, writes out a new MTZ file and repeats all the checks on the output MTZ file. It is essential to check the log file to ensure that the output file passes all the tests; if it doesn't the faulty columns indicated by the program should not be used.
The assumed default file labels are as follows:
For Fo & sigma(Fo): 'FOSC' & 'SIGFOSC' (Buster), or 'F' & 'SIGF'
(Refmac & Sigmaa).
The Fo and sigma(Fo) file labels may also be 'F_' and 'SIGF_'
followed by the same arbitrary string (up to 24 characters). In
this case it's not necessary to specify an FLABEL string.
For Fc & phic: 'FC_ALL' & 'PHIC_ALL' (Refmac only), or 'FC' & 'PHIC' (all). However, note that use of the standard FC and PHIC columns from Refmac will cause most of the tests to fail; it is essential that the program uses the default 'FC_ALL' & 'PHIC_ALL' columns (i.e. containing the total contribution to Fc including that from the bulk solvent).
For '2Fo-Fc' and associated phase: '2FOFCWT' & 'PH2FOFCWT' (Buster), or 'FWT' & 'PHWT' (Refmac & Sigmaa).
For 'Fo-Fc' and associated phase: 'FOFCWT' & 'PHFOFCWT' (Buster), or 'DELFWT' & 'PHDELWT' (Refmac & Sigmaa).
For FOM: 'FOM' (Buster & Refmac) or 'WCMB' (Sigmaa).
For free R flag: 'FREE' or 'FreeR_flag'.
mtzfix HKLIN input.mtz HKLOUT output.mtzThis example assumes labels FP/SIGFP for Fo & defaults for everything else:
mtzfix FLABEL FP HKLIN input.mtz HKLOUT output.mtz