THE AB-INITIO CRYSTAL-STRUCTURE SOLUTION OF PROTEINS BY DIRECT-METHODS .1. FEASIBILITY

Traditional direct methods based on the tangent formula and/or on Sayre's equation cannot solve ab initio the large majority of protein crystal structures [Giacovazzo, Guagliardi, Ravelli & Siliqi (1994). Z. Kristallogr. 209, 136-142]. Indeed, the amount of information available leads to a signal-to-noise ratio close to unity; consequently, the correct solution, even if attained, cannot be recognized among the trial solutions. Attention is here focused onto the case in which diffraction data of one isomorphous derivative are additionally available. It is shown that in such a case direct ab initio solution of protein structures is feasible. Tests based on calculated diffraction data suggest the procedure to follow for a possible success.