Macrocycles measured

The old melds with the new

Direct head-to-head comparisons of computational techniques are rare, which is why this study exploring the use of widely used conformational search algorithms for reproducing flexible macrocycle structures is valuable. The authors compare several standard conformational search protocols (Schrödinger, OpenEye, MOE) and point out the strengths and weaknesses of each in reproducing known crystal structures, global energy minima, number of search steps, impact of polarity of medium, intramolecular hydrogen bonds, side chain vs core flexibility and other key aspects of important macrocycle drugs (antibiotics and HCV protease inhibitors). They point out that the radius of gyration (which effectively measures flexibility of a molecule around the mean) and PSA are much more valuable metrics for comparing results than more standard measures like RMSD.

On a personal note this paper was interesting to me as well, in part because I remember successfully employing low-mode MC conformational and search and Rgyr to explore the conformational space of similarly-sized macrocycles almost fifteen years ago. The field has significantly advanced since then, but it's reassuring to know that many of the old methods and assessment techniques still work well. More than anything else, these algorithms are a testament to the enduring robustness and value of force fields like MMFF94 and OPLS whose foundations were laid years ago using careful, meticulous parameterization and testing. Science truly stands on the shoulders of giants.