Multistep Reaction Based De Novo Drug Design: Generating Synthetically Feasible Design Ideas

We describe a "multistep reaction driven" evolutionary algorithm approach to de novo molecular design. Structures generated by the approach include a proposed synthesis path intended to aid the chemist in assessing the synthetic feasibility of the ideas that are generated. The methodology is independent of how the design ideas are scored, allowing multicriteria drug design to address multiple issues including activity at one or more pharmacological targets, selectivity, physical and ADME properties, and off target liabilities; the methods are compatible with common computer-aided drug discovery "scoring" methodologies such as 2D- and 3D-ligand similarity, docking, desirability functions based on physiochemical properties, and/or predictions from 2D/3D QSAR or machine learning models and combinations thereof to be used to guide design. We have performed experiments to assess the extent to which known drug space can be covered by our approach. Using a library of 88 generic reactions and a database of ∼20 000 reactants, we find that our methods can identify "close" analogs for ∼50% of the known small molecule drugs with molecular weight less than 300. To assess the quality of the in silico generated synthetic pathways, synthesis chemists were asked to rate the viability of synthesis pathways: both "real" and in silico generated. In silico reaction schemes generated by our methods were rated as very plausible with scores similar to known literature synthesis schemes.

Virtual screening for R-groups, including predicted pIC50 contributions, within large structural databases, using Topomer CoMFA

Multiple R-groups (monovalent fragments) are implicitly accessible within most of the molecular structures that populate large structural databases. R-group searching would desirably consider pIC50 contribution forecasts as well as ligand similarities or docking scores. However, R-group searching, with or without pIC50 forecasts, is currently not practical. The most prevalent and reliable source of pIC50 predictions, existing 3D-QSAR approaches, is also difficult and somewhat subjective. Yet in 25 of 25 trials on data sets on which a field-based 3D-QSAR treatment had already succeeded, substitution of objective (canonically generated) topomer poses for the original structure-guided manual alignments produced acceptable 3D-QSAR models, on average having almost equivalent statistical quality to the published models, and with negligible effort. Their overall pIC50 prediction error is 0.805, calculated as the average over these 25 topomer CoMFA models in the standard deviations of pIC50 predictions, derived from the 1109 possible "leave-out-one-R-group" (LOORG) pIC50 contributions. (This novel LOORG protocol provides a more realistic and stringent test of prediction accuracy than the customary "leave-out-one-compound" LOO approach.) The associated average predictive r(2) of 0.495 indicates a pIC50 prediction accuracy roughly halfway between perfect and useless. To assess the ability of topomer-CoMFA based virtual screening to identify "highly active" R-groups, a Receiver Operating Curve (ROC) approach was adopted. Using, as the binary criterion for a "highly active" R-group, a predicted pIC50 greater than the top 25% of the observed pIC50 range, the ROC area averaged across the 25 topomer CoMFA models is 0.729. Conventionally interpreted, the odds that a "highly active" R-group will indeed confer such a high pIC50 are 0.729/(1-0.729) or almost 3 to 1. To confirm that virtual screening within large collections of realized structures would provide a useful quantity and variety of R-group suggestions, combining shape similarity with the "highly active" pIC50, the 50 searches provided by these 25 models were applied to 2.2 million structurally distinct R-group candidates among 2.0 million structures within a ZINC database, identifying an average of 5705 R-groups per search, with the highest predicted pIC50 combination averaging 1.6 log units greater than the highest reported pIC50s.

AllChem: generating and searching 1020 synthetically accessible structures

AllChem is a system that is intended to make practical the generation and searching of an unprecedentedly vast number ( approximately 10(20)) of synthetically accessible and medicinally relevant structures. Also, by providing possible synthetic routes to a structure along with its design rationale, AllChem encourages simultaneous consideration of both costs and benefits during each lead discovery and optimization decision, thereby promising to be effective with synthetic chemists among its primary users. AllChem is still under intensive development so the following initial description necessarily has more the character of an interim progress report than of a finished research publication.

Balancing focused combinatorial libraries based on multiple GPCR ligands

G-Protein coupled receptors (GPCRs) are important targets for drug discovery, and combinatorial chemistry is an important tool for pharmaceutical development. The absence of detailed structural information, however, limits the kinds of combinatorial design techniques that can be applied to GPCR targets. This is particularly problematic given the current emphasis on focused combinatorial libraries. By linking an incremental construction method (OptDesign) to the very fast shape-matching capability of ChemSpace, we have created an efficient method for designing targeted sublibraries that are topomerically similar to known actives. Multi-objective scoring allows consideration of multiple queries (actives) simultaneously. This can lead to a distribution of products skewed towards one particular query structure, however, particularly when the ligands of interest are quite dissimilar to one another. A novel pivoting technique is described which makes it possible to generate promising designs even under those circumstances. The approach is illustrated by application to some serotonergic agonists and chemokine antagonists.

OptDesign: extending optimizable k-dissimilarity selection to combinatorial library design

Optimizable k-dissimilarity (OptiSim) selection entails drawing a series of subsamples of size k from a population and choosing the "best" candidate from each such subsample for inclusion in the selection set. By varying the size of the subsample, one can control the balance between representativeness and diversity in the selection set obtained. In the original formulation, a uniform random sampling from among valid candidates was used to draw the subsamples from a single target population. Here we describe in detail two key modifications that serve to extend the OptiSim methodology to vector selection for interdependent variables, specifically as applied to the design of combinatorial sublibraries. The first modification involves pivoting between variables: subsamples are drawn from each reagent pool in turn, with the viability of each candidate being evaluated in isolation as well as in terms of the products it will produce from complementary reagents already selected. The filters applied may be static or dynamic in nature, with molecular weight and hydrophobicity being examples of the former and structural diversity with respect to reagents already selected being an example of the latter. The second key modification is adding the ability to bias the selection of candidate reagents for inclusion in the subsamples. Taken together, these modifications support the efficient generation of multiblock and other sparse matrix designs that are both representative and diverse, and for which "backfilling" of designs edited to remove undesirable reagents or products is straightforward. The method is intrinsically fast and efficient, since enumeration of the full combinatorial is not required- only those candidates actually considered for inclusion need be evaluated. Moreover, because the subsample selection step is separate from the diversity-based selection of the "best" candidate, incorporating such bias in favor of a competing criterion such as low price provides a "natural," nonparametric mechanism for generating designs that are likely to be "good" in a double-objective, Pareto sense.

Visualizing substructural fingerprints

Substructural fingerprints have proven very useful for chemical library and diversity analysis, but their high dimensionality makes them poorly suited to principal components analysis and to standard nonlinear mapping methods. By using a combination of optimizable K-dissimilarity selection (OptiSim) and a modified stress function that suppresses effects of distances that fall beyond a characteristic horizon, it is possible to relax principal components analysis coordinates into more consistently meaningful projections from fingerprint space into two dimensions. The nonlinear maps so obtained are useful for characterizing combinatorial libraries, for comparing sublibraries, and for exploring the distribution of biological properties across structural space.