MOLECULAR SPELUNKING: AutoDock Vina evaluates a receptor’s entire binding pocket at once to find a docking site (lowest-energy binding) for a ligand of interest. The deepest part of the pocket (the actual binding site for auxin shown at the bottom of (1) naturally has a low energy requirement, and this is where, in the case of tryptophan (2), the software suggests a docking site. But by forcing the software to move in incremental steps—starting at the mouth of the pocket and moving inward—TomoDock finds an interaction of tryptophan with residues farther up the pocket (3) that prevents deeper entry. In the case of auxin, however, TomoDock finds the same binding site as that found using AutoDock Vina (step-wise progress of TomoDock shown in (1).© GEORGE RETSECK

Auxins are a family of small-molecule hormones that control growth and development processes in plants. They are also components of widely used herbicides. In a drive to extend the agricultural and horticultural applications of these hormones, scientists are attempting to design new synthetic auxins. But to do so, they must understand the nitty-gritty of how an auxin molecule binds to it receptor, says Richard Napier of the University of Warwick in the U.K.

Napier’s team uses docking software to simulate auxin binding. But there’s a problem: the software also allows molecules to dock that, Napier says, are known not to bind in reality—such as auxin’s close relative tryptophan. “Getting false positives out of docking [analyses] is absolutely part of the deal,” he says. “Docking is not a perfect science.”

To reduce such permissiveness, Napier and his colleagues have written additional computational code for a popular docking program, AutoDock Vina. For receptors with deep binding-site pockets (like that of the auxin receptor), the new code mimics the molecule’s natural passage by searching for the docking site in a sequence of 0.1-nanometer steps.

Drawing an analogy with a cave, Ning Zheng of the University of Washington says, “The conventional [docking] method just looks at whether a child or adult can be accommodated by the cave interior, but if that cave is separated from the outside by, let’s say, a narrow cleft, then . . . maybe it turns out the adult is too big to pass [through].”

Restriction of a molecule’s access may not be due to size, but to interactions with residues on the “cave” walls as it enters—which is the case for tryptophan, Napier’s team has now discovered. Without simulating a molecule’s passage, as per the new method, called TomoDock, such interactions can be missed, Napier says. (Open Biology, 6:160139, 2016) 

DOCKING PROGRAM BINDING SITE EVALUATION
 
APPLICABILITY ACCESS
AutoDock Vina A 3-D cuboidal area encompassing the entire binding pocket of a receptor All ligand-receptor interactions Free to download from vina.scripps.edu
TomoDock (with AutoDock Vina) A 3-D cuboidal area big enough to encompass the entire binding pocket, but which is initially positioned at the mouth of the binding pocket and then moved stepwise into the pocket in small increments Specifically for receptors with deep binding pockets or transport proteins in which molecules of interest pass though a channel Free to download from www2.warwick.ac.uk/fac/cross_fac/
complexity/people/staff/delgenio/tomocode