Table 6. Decision-making accuracies in choosing the solution with the best experimental or density-modified map.

The percentage of correct predictions of best maps is the percentage of cases in which the solution with the highest value of the quality measure has a map correlation coefficient with the corresponding model map within 0.02 of that of the best obtained for any solution for that structure. The analysis is based on 372 sets of structure factors and associated maps obtained from 149 data sets as in Fig. 1 ▶, selecting the top-ranked 2–6 solutions and carrying out density modification with RESOLVE (Terwilliger, 2000 ▶) to yield density-modified maps. A model was built into each density-modified map using a rapid method for building helices and strands. If the value of the map–model correlation was less than 0.35, then the building procedure was repeated with a standard cycle of building using the methods in the PHENIX AutoBuild wizard (Terwilliger et al., 2008 ▶) and the value of the map–model correlation from the full standard procedure was used. Only structures for which at least one model–map correlation was at least 0.20 are included in the analysis. The worst error in identification of the best maps is the largest value of the difference between the correlation coefficient of the best map with the corresponding model map and that of the map with the highest value of the quality measure.

	Percentage of correct predictions of best maps		Worst error in identification of best maps
Quality measure	Experimental maps	Density-modified maps	Experimental maps	Density-modified maps
Bayesian estimate using skew and r2RMS of experimental map	91	88	0.29	0.58
Map-model correlation for model built into density-modified map	87	92	0.40	0.26