Bradley P. Feuston

Random Forest: A Classification and Regression Tool for Compound Classification and QSAR Modeling

A new classification and regression tool, Random Forest, is introduced and investigated for predicting a compounds quantitative or categorical biological activity based on a quantitative description of the compounds molecular structure. Random Forest is an ensemble of unpruned classification or regression trees created by using bootstrap samples of the training data and random feature selection in tree induction. Prediction is made by aggregating (majority vote or averaging) the predictions of the ensemble. We built predictive models for six cheminformatics data sets. Our analysis demonstrates that Random Forest is a powerful tool capable of delivering performance that is among the most accurate methods to date. We also present three additional features of Random Forest: built-in performance assessment, a measure of relative importance of descriptors, and a measure of compound similarity that is weighted by the relative importance of descriptors. It is the combination of relatively high prediction accuracy and its collection of desired features that makes Random Forest uniquely suited for modeling in cheminformatics.

Comparison of knowledge-based and distance geometry approaches for generation of molecular conformations.

A knowledge-based approach for generating conformations of molecules has been developed. The method described here provides a good sampling of the molecules conformational space by restricting the generated conformations to those consistent with the reference database. The present approach, internally named et for enumerate torsions, differs from previous database-mining approaches by employing a library of much larger substructures while treating open chains, rings, and combinations of chains and rings in the same manner. In addition to knowledge in the form of observed torsion angles, some knowledge from the medicinal chemist is captured in the form of which substructures are identified. The knowledge-based approach is compared to Blaney et al.s distance geometry (DG) algorithm for sampling the conformational space of molecules. The structures of 113 protein-bound molecules, determined by X-ray crystallography, were used to compare the methods. The present knowledge-based approach (i) generates conformations closer to the experimentally determined conformation, (ii) generates them sooner, and (iii) is significantly faster than the DG method.

Potent and selective HIV-1 ribonuclease H inhibitors based on a 1-hydroxy-1,8-naphthyridin-2(1H)-one scaffold

Optimization studies using an HIV RNase H active site inhibitor containing a 1-hydroxy-1,8-naphthyridin-2(1H)-one core identified 4-position substituents that provided several potent and selective inhibitors. The best compound was potent and selective in biochemical assays (IC(50)=0.045 μM, HIV RT RNase H; 13 μM, HIV RT-polymerase; 24 μM, HIV integrase) and showed antiviral efficacy in a single-cycle viral replication assay in P4-2 cells (IC(50)=0.19 μM) with a modest window with respect to cytotoxicity (CC(50)=3.3 μM).

Identification of metabotropic glutamate receptor antagonists using an automated high-throughput screening system

Antagonists to the human metabotropic glutamate receptor subtype 5a(mGluR(5a)) have been implicated as potential therapeutics for the treatment of a variety of nervous system disorders, including pain, anxiety, and Parkinsons disease. To discover novel antagonists to the mGluR(5a), a functional assay measuring agonist-induced intracellular calcium release was developed. The assay was used for the high-throughput screening of a large collection of compounds in single wells using a fully automated robotic platform. Primary high-throughput screening hits were subjected to a combination of data analysis and counterscreening assays to identify several compounds with both efficacy and selectivity for the metabotropic glutamate receptor target.

Targeted purification development enabled by computational biophysical modeling

Chromatographic and non‐chromatographic purification of biopharmaceuticals depend on the interactions between protein molecules and a solid–liquid interface. These interactions are dominated by the protein–surface properties, which are a function of protein sequence, structure, and dynamics. In addition, protein–surface properties are critical for in vivo recognition and activation, thus, purification strategies should strive to preserve structural integrity and retain desired pharmacological efficacy. Other factors such as surface diffusion, pore diffusion, and film mass transfer can impact chromatographic separation and resin design. The key factors that impact non‐chromatographic separations (e.g., solubility, ligand affinity, charges and hydrophobic clusters, and molecular dynamics) are readily amenable to computational modeling and can enhance the understanding of protein chromatographic. Previously published studies have used computational methods such as quantitative structure–activity relationship (QSAR) or quantitative structure–property relationship (QSPR) to identify and rank order affinity ligands based on their potential to effectively bind and separate a desired biopharmaceutical from host cell protein (HCP) and other impurities. The challenge in the application of such an approach is to discern key yet subtle differences in ligands and proteins that influence biologics purification. Using a relatively small molecular weight protein (insulin), this research overcame limitations of previous modeling efforts by utilizing atomic level detail for the modeling of protein–ligand interactions, effectively leveraging and extending previous research on drug target discovery. These principles were applied to the purification of different commercially available insulin variants. The ability of these computational models to correlate directionally with empirical observation is demonstrated for several insulin systems over a range of purification challenges including resolution of subtle product variants (amino acid misincorporations). Broader application of this methodology in bioprocess development may enhance and speed the development of a robust purification platform.

Dynamic control of allosteric antagonism of leukocyte function antigen-1 and intercellular adhesion molecule-1 interaction.

Leukocyte function associated antigen‐1 (LFA‐1) plays a critical role in T cell migration and has been recognized as a therapeutic target for immune disorders. Several classes of small molecule antagonists have been developed to block LFA‐1 interaction with intercellular adhesion molecule‐1 (ICAM‐1). Recent structural studies show that the antagonists bind to an allosteric site in the I‐domain of LFA‐1. However, it is not yet clear how these small molecules work as antagonists since no significant conformational change is observed in the I‐domain–antagonist complex structures. Here we present a computational study suggesting how these allosteric antagonists affect the dynamics of the I‐domain. The lowest frequency vibrational mode calculated from an LFA‐1 I‐domain structure shows large scale “coil‐down” motion of the C‐terminal α7 helix, which may lead to the open form of the I‐domain. The presence of an allosteric antagonist greatly reduces this motion of the α7 helix as well as other parts of the I‐domain. Thus, our study suggests that allosteric antagonists work by eliminating breathing motion that leads to the open conformation of the I‐domain. Proteins 2006.

Web Enabling Technology for the Design, Enumeration, Optimization and Tracking of Compound Libraries

Motivated by the need to augment Mercks in-house small molecule collection, web-based tools for designing, enumerating, optimizing and tracking compound libraries have been developed. The path leading to the current version of this Virtual Library Tool Kit (VLTK) is discussed in context of the (then) available commercial offerings and the constraints and requirements imposed by the end users. Though the effort was initiated to simplify the tasks of designing novel, drug-like and diverse compound libraries containing between 2K-10K unique entities, it has also evolved into a powerful tool for outsourcing syntheses as well as lead identification and optimization. The web tool includes components that select reagents, analyze synthons, identify backup reagents, enumerate libraries, calculate properties, optimize libraries and finally track the synthesized compounds through biological assays. In addition to accommodating project specific designs and virtual 3D library scanning, the application includes tools for parallel synthesis, laboratory automation and compound registration.

Scoring of KDR Kinase Inhibitors: Using Interaction Energy as a Guide for Ranking

SummaryWithin a congeneric series of ATP-competitive KDR kinase inhibitors, we determined that the IC50 values, which span four orders of magnitude, correlated best with the calculated ligand-protein interaction energy using the Merck Molecular Force Field (MMFFs(94)). Using the ligand-protein interaction energy as a guide, we outline a workflow to rank order virtual KDR kinase inhibitors prior to synthesis. When structural information of the target is available, the ability to score molecules a priori can be used to rationally select reagents. Our implementation allows one to select thousands of readily available reagents, enumerate compounds in multiple poses and score molecules in the active site of a protein within a few hours. In our experience, virtual library enumeration is best used when a correlation between computed descriptors/properties and IC50 or Ki values has been established.

Reagent Selector: using Synthon Analysis to visualize reagent properties and assist in combinatorial library design.

Reagent Selector is an intranet-based tool that aids in the selection of reagents for use in combinatorial library construction. The user selects an appropriate reagent group as a query, for example, primary amines, and further refines it on the basis of various physicochemical properties, resulting in a list of potential reagents. The results of this selection process are, in turn, converted into synthons: the fragments or R-groups that are to be incorporated into the combinatorial library. The Synthon Analysis interface graphically depicts the chemical properties for each synthon as a function of the topological bond distance from the scaffold attachment point. Displayed in this fashion, the user is able to visualize the property space for the universe of synthons as well as that of the synthons selected. Ultimately, the reagent list that embodies the selected synthons is made available to the user for reagent procurement. Application of the approach to a sample reagent list for a G-protein coupled receptor targeted library is described.