Abstract
Template CoMFA methodologies extend topomer CoMFA by allowing user-designated templates, for example the experimental receptor-bound conformation of a prototypical ligand, to help determine the alignment of training and test set structures for 3D-QSAR. The algorithms that generate its new structural modality, template-constrained topomers, are described. Template CoMFA’s resolution of certain topomer CoMFA concerns, by providing user control of topological consistency and structural acceptability, is demonstrated for sixteen 3D-QSAR training sets, in particular the Selwood dataset.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Lead optimization costs, per new drug introduction, are the highest of all, exceeding those of Phase II and III development because, being earlier, they generate more dead-ends and tie up capital for longer. More specifically, lead optimization accounts for 17 % of total R&D cost and around 50 % of discovery cost, and may be the 3rd largest opportunity area for overall R&D cost reduction.
Challenges for its chief competitor in practice, binding free energy calculation, are most recently discussed by Stouch [4]
Patents are pending on template-constrained topomers and their applications.
Large “# component” values may raise concerns about over-fitting, especially when accompanied by unreasonably low “SDEP” values. However, from PLS over-fitted and unstable models are much less of a practical risk than from other common algorithms such as multiple regression, because PLS operates on blocks of descriptors rather than individual columns. The usual effects of additional components on a PLS model are increasingly minor refinements, seldom having any effect on overall “statistical significance”. Therefore, in the standard topomer CoMFA implementation as used in these studies, during leave-one-out cross-validation, component extraction ends only when the resulting SDEP value first increases. Of course the analyst may then truncate the “#components” to a smaller value, but in these studies, such a necessarily subjective decision seemed inappropriate.
References
Paul SM, Mytelka DS, Dunwiddie CT, Persinger CC, Munos BH, Lindberg SR, Schacht AL (2010) Nat Rev Drug Disc 9:203–214
Cramer RD, Cruz P, Stahl G, Curtiss WC, Campbell B, Masek BB, Soltanshahi F (2008) J Chem Inf Model 48:2180–2195
Doweyko A (2007) J Comp Aided Drug Des 18:587–596
Stouch TR (2012) J Comp Aided Drug Des 26:125–134
Cramer RD, Clark RD, Patterson DE, Ferguson AM (1996) J Med Chem 39:3060–3069
Jilek RJ, Cramer RD (2004) J Chem Inf Comp Sci 44:1221–1227
Cramer RD, Jilek RJ, Guessregen S, Clark SJ, Wendt B, Clark RD (2004) J Med Chem 47(6777):6791
Cramer RD (2003) J Med Chem 46:374–389
Cramer RD (2012) J Comp Aided Drug Des 25:197–201
Selwood DL, Livingston DJ, Comley JCW, O’Dowd AB, Hudson AT, Jackson P, Jandu KS, Rose VS, Stables JM (1990) J Med Chem 33:136–142
Nicoletti O, Gillet VJ, Fleming PJ, Green DVS (2002) J Med Chem 45:5069–5080
Kubinyi H (1994) Quant Struct Act Relat 13:285–294
So S-S, Karplus M (1996) J Med Chem 39:1521–1540
Wendt B, Cramer RD (2008) J Comp Aided Drug Des 22:541–551
Jain AN (2004) J Med Chem 47:941–961
Maggiora GM (2006) J Chem Inf Model 46:1535
Taleb NN (2007) The Black Swan: the impact of the highly Improbable, Random House, ISBN 978-1-4000-6351-2
Acknowledgment
It is a great pleasure to thank Bernd Wendt for calling attention to the didactic features of the Selwood data set.
Author information
Authors and Affiliations
Corresponding author
Electronic supplementary material
All templates and training sets referenced in the Results section.
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Cramer, R.D. R-group template CoMFA combines benefits of “ad hoc” and topomer alignments using 3D-QSAR for lead optimization. J Comput Aided Mol Des 26, 805–819 (2012). https://doi.org/10.1007/s10822-012-9583-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10822-012-9583-9