datset isomorphs to be included

limset heavy atoms to be included

datscl scale isomorph data to parent data

ref parameters to be refined

noref turn off parameter refinement

bins statistics grouping

stat statistics to be calculated and printed

nostat turn off statistics to be calculated and printed

mvfc¹MVFC phase determination

fazinintroduce known phases (currently for MVFC method only)

fazprb¹modified Blow-Crick phase determination

eiso isomorphous closure errors

eano anomalous closure errors

plthkl probability profile plotting

newtonadjustment of most probable phase by Newton's method

cycle number of refinement cycles

1 If neither appears no phase determination will be made.

MIR 1 phase refinement control

block block diagonal

full full matrix Bricogne method (not implemented)

none no phase reference

2 list frequency on last cycle no listing

3 reflection listing control basic items

native basic + expanded list for parent, only

all basic + expanded list for all data sets

4 output fazprb closure errors on last cycle no

closure list closure errors

datset 1 data set name (12 characters)

2 data used for phasing

both use both isomorphs and anomalous

isom use isomorphs only

anom use anomalous only

none do not phase

3 data used for refinement (same controls as field 2)

atoms 1 data set name

2- atom labels of atoms to be used

limset 1 dataset name of isomorph

2 min sinθ/λ for hkl of this isomorph 0

3 max sinθ/λ for hkl of this isomorph bdf or 2

datscl 1 dataset name of isomorph

2 value to scale isomorph to parent bdf or 1

3 value of ΔB to scale isomorph to parent bdf or 0

4 observational errors (σ's) for each reflection 0.05

If an isomorph does not have these available, the program calculates them as follows:

σ²= (1 + C*F)²

where C is analogous to the instrumental instability factor used when calculating observational errors from diffractometer data.

ref 1 atom label, dataset name or 'all' as global signal

noref 2- referenced parameter as mnemonic

x y z atomic coordinates

u u11 u22 u33 u12 u13 u23 thermal parameters

pp ppa pop. param (real+anomalous)

k c overall scale and ΔU

bins 1 number of statistics bins (max. 12)¹ 5

2 number of statistics bins (max. 12)¹ 3

3 statistics listing control for each cycle except last²

all list all categories of statistics³

centric list centric statistics only

acentric list acentric statistics only

4 statistics listing control for last cycle³ (same controls as in field 3)

centric

1 The total number of bins is the product of the number of ranges and the bins. The total may be as few as 1 bin or as many as 50. To produce reasonable statistics a minimum of 100 reflections should appear in each bin and preferably several times that number.

2 Statistics output can be quite copious. In addition to the stat and nostat lines below, fields 3 and 4 offer an opportunity to limit the statistic listing to centric, acentric or all.

3 As the isomorphous closure errors in fazprb mode are updated from the centric values (if there were any centric measurements), these are also printed under the all option.

stat codes defining statistical analyses (see description in Ref. Manual)

nostat fomhist figure-of-merit histogram

fom figure-of-merit by and

rcullis Cullis R-factor

rkraut Kraut R-factor

rfazpwr R phasing power; weighted closure error over

weighted heavy atom model

wteiso RMS weighted isomorphous closure error (on F)

wteano RMS weighted anomalous closure error (on F)

number number of reflections

fazdif phase differences between parent & heavy atom

model

all all statistics

lstcyc listed codes only apply to last cycle

If no fields follow this mnemonic, all statistics which have not already been requested are printed on the last cycle. Only those codes listed on the lstcyc line are printed on the last cycle.

for FAZPRB mode only:

parvar parent variance

eiso RMS isomorphous closure error

eisoadj RMS isomorph. closure errors adjusted for variance

isodif RMS isomorphous differences

hamodel RMS heavy atom model

jntvar RMS joint variance of parent and isomorph

ermsano RMS anomalous closure errors

eanoadj RMS anom. closure errors adjusted for variances

anodif RMS anomalous differences

signs agreement of obs. and calc. anomalous diffrences

The statistical output can be copious. To save trees but not the movement of electrons, the stat and nostat lines set the print priority of the desired statistics to 3 and all other statistics to 5 (i.e. only output if reset pstat 5 is entered before MIR). A very small group of statistics always has priority 2 for those who wish to monitor the program while it is being run in the "interactive" mode. These statistics include the average phase change from the previous cycle, a summary providing the overall figure of merit, and for each isomorph the overall R Cullis, R Kraut and R phasing power. In fazprb mode, ermsiso and ermsano are also calculated.

The default is to print to all statistics on the last cycle and the following statistics on all cycles: fom, rcullis, rkraut, rfazpwr and in fazprb mode ermsiso, eisoadj, ermsano and eanoadj. Statistics relating to anomalous information are printed only if there were observations.

mvfc 1 phase increment (max. 180^o) 10

2 weak reflection cutoff ratio F/γ(δ) 2.5

3 phase refinement criterion selection 1

4 shift factor for phases (max. 10) 1

5 phase threshold to accept bimodal SIR soln. (max. 179^o) 179

6 goodness-of-fit to initialize FOM bdf or 20

7 signal to not use bdf phases, enter newphases use bdf

fazin Option Code

initial phase set init n 1

restraining phase set comb n 1

fazprb 1 phase incr. for integrating refln. probability profile 5

(min. 1^o; max. 180^o; recommended max. 12^o)

2 signal to control refinement and statistics

most use the most probable phases

best use the 'best' phases

eiso and eano (for fazprb mode only)

1 dataset ID

2 number of bins reserved for closure errors NO default 0 implies all bins used

3- closure errors

The number of closure errors must either be 1 or equal to the current number of ranges (see bins line). If only 1 closure error is encountered, all ranges within the indicated bins (field 2) are replaced with that value. Otherwise there is corresponding replacement. The usual default value for the closure errors are those on the bdf. If the bdf has not been used by MIR, default for isomorphous closure errors is 1000 and anomalous closure errors is 592. If the number of ranges or bins is changed (see bins line) and there are closure errors on the binary data file, new closure errors are expected to be supplied from the input stream.

plthkl (for fazprb mode only)

1-3 hkl indices of the first of a group of reflections to be plotted

4 number of reflections in group to be plotted (max. 100) 10

This line provides a facility for line printer plots of the phase probability profile for a group of reflections. This operates only on the last cycle of the program and prints at priority 4, and is intended primarily as a teaching tool to study the effects of various parameters on the phase distribution of selected reflections. Each plot takes a page of output so this feature should be used sparingly. The program will not perform this operation when the listing of every nth reflection has been requested on the MIR line.

newton 1 centroid minimum level for centro reflections 0.5

2 centroid minimum level for noncentro reflections 0.8

cycle 1 number of refinement cycles (max. 10) 1