Practical applications of deep learning to impute heterogeneous drug discovery data
This article outlines practical applications of deep learning on drug discovery data. It introduces some of the research behind our…
Structure- and ligand-based virtual screening on DUD-E+: performance dependence on approximations to the binding pocket
Using the DUD-E+ benchmark, we explore the impact of using a single protein pocket or ligand for virtual screening compared with using ensembles of alternative pockets, ligands, and sets thereof.
Predicting pKa using a combination of quantum mechanical and machine learning methods
This study aimed to create a model for predicting pKa using a semi-empirical quantum mechanics (QM) approach combined with machine learning (ML).
Electrostatic-field and surface-shape similarity for virtual screening and pose prediction
We introduce a new method for rapid computation of 3D molecular similarity that combines electrostatic field comparison with comparison of molecular surface-shape and directional hydrogen-bonding preferences (called “eSim”).
N- and S-oxidation model of the flavin-containing monooxygenases
Introduction Existing computational models of drug metabolism are heavily focused on predicting oxidation by cytochrome P450 (CYP) enzymes, because of…
Complex macrocycle exploration: parallel, heuristic, and constraint-based conformer generation using ForceGen
ForceGen is both faster and more accurate than the best of all tested methods on a very large, independently curated benchmark of 2859 PDB ligands. In this study, the primary results are on macrocycles, including results for 431 unique examples from four separate benchmarks.
Imputation of assay bioactivity data using deep learning
This article describes a novel deep learning neural network method and its application for the imputation of bioactivity data, such…
A novel scoring profile for the design of antibacterials active against gram-negative bacteria
Introduction The increasing occurrence of multidrug-resistant bacteria is one of the major global threats to human health. Design of new…
Quantitative surface field analysis: learning causal models to predict ligand binding affinity and pose
We introduce the QuanSA method for inducing physically meaningful field-based models of ligand binding pockets based on structure-activity data alone.
WhichP450: a multi-class categorical model
This paper describes the underlying methods and validation of the WhichP450 model, which predicts the most likely Cytochrome P450 isoforms…
ForceGen 3D structure and conformer generation: from small lead-like molecules to macrocyclic drugs
We introduce the ForceGen method for 3D structure generation and conformer elaboration of drug-like small molecules.
Practical applications of matched series analysis
This paper, co-authored with our colleagues at NextMove Software, explores applications of Matched Series Analysis within StarDrop’s Nova module to…
Avoiding missed opportunities by analysing the sensitivity of our decisions
This peer-reviewed article, published in the Journal of Medicinal Chemistry, describes how identifying sensitive criteria can highlight new avenues for exploration, and assist us in avoiding missed opportunities
Predicting regioselectivity and lability of P450 metabolism
This article describes the underlying methods, validation and example applications of the most recent models of Cytochrome P450 metabolism in…
Extrapolative prediction using physically-based QSAR
Surflex-QMOD integrates chemical structure and activity data to produce physically-realistic models for binding affinity prediction.
The challenges of making decisions using uncertain data
This peer-reviewed paper discusses the challenges of using uncertain experimental data to make confident decisions on the selection of compounds.…
Knowledge-guided docking: accurate prospective prediction of bound configurations of novel ligands using Surflex-Dock
We present an approach that uses structural information known prior to a particular cutoff-date to make predictions on ligands whose bounds structures were determined later. The knowledge-guided docking protocol was tested on a set of ten protein targets using a total of 949 ligands.
Breaking free from chemical spreadsheets
This article explores the benefits of a more intuitive and flexible approach to viewing and interacting with drug discovery data,…
Chemical and protein structural basis for biological crosstalk between PPARα and COX enzymes
Here we present an analysis of novel drug/target predictions, focusing on those that were not obvious based on known pharmacological crosstalk.
