For what it's worth, I'd try to get as much as possible out of the experimental phases before going on to phased MR.
> I have SeMet MAD data to 3.6A that gives decent looking anomalous > difference peaks, looks stable in mlphare, and produces solvent > flattened maps to 2.8A in DM that look like the density might be a > sensible shape - wrt solvent gaps etc - but not interpretable so far (I > will be trying phasing in SHARP, CNS etc). One thing to try would be the bp3/pirate combination for phasing/density modification. With some datasets, this works very well (in my hands, not with all datasets). Another (which you've probably already tried) would be to check the MAD maps after solvent flattening, but before phase extension, as well as comparing the MAD maps to SAS maps. It's probably also worth checking the dispersive difference peaks, and cross checking them with the anomalous difference peaks. You should get the same peaks out, but it'll let you guess as the relative phasing signal for dispersive vs anomalous (but so will comparing MAD vs SAS). > Any advice gratefully received. Hopefully I'm not giving you too many suggestions you've already tried. Pete
