A5040409076- Computational Drug Discovery Methods
- Microbial Natural Products and Biosynthesis
- Machine Learning in Materials Science
- Protein Structure and Dynamics
- Chemical Synthesis and Analysis
- Cell Image Analysis Techniques
- Design Education and Practice
- Bioinformatics and Genomic Networks
University of Oxford
2021-2023
Oxford Research Group
2023
A novel deep generative model combines convolution and graph neural networks to allow 3D-aware molecular design. We show how 3D pharmacophoric information can be incorporated into models apply our both linker R-group
Over the past few years, many machine learning-based scoring functions for predicting binding of small molecules to proteins have been developed. Their objective is approximate distribution which takes two as input and outputs energy their interaction. Only a function that accounts interatomic interactions involved in can accurately predict affinity on unseen molecules. However, make predictions based data set biases rather than an understanding physics binding. These perform well when...
Many recently proposed structure-based virtual screening models appear to be able accurately distinguish high affinity binders from non-binders. However, several recent studies have shown that they often do so by exploiting ligand-specific biases in the dataset, rather than identifying favourable intermolecular interactions input protein-ligand complex. In this work we propose a novel approach for assessing extent which machine learning-based are identify functional groups responsible...
Abstract Many recently proposed structure-based virtual screening models appear to be able accurately distinguish high affinity binders from non-binders. However, several recent studies have shown that they often do so by exploiting ligand-specific biases in the dataset, rather than identifying favourable intermolecular interactions input protein-ligand complex. In this work we propose a novel approach for assessing extent which machine learningbased are identify functional groups...
Abstract Generative models have increasingly been proposed as a solution to the molecular design problem. However, it has proved challenging control process or incorporate prior knowledge, limiting their practical use in drug discovery. In particular, generative methods made limited of three-dimensional (3D) structural information even though this is critical binding. This work describes method such and demonstrates benefit doing so. We combine an existing graph-based deep model, DeLinker,...
ABSTRACT Over the last few years, many machine learning-based scoring functions for predicting binding of small molecules to proteins have been developed. Their objective is approximate distribution which takes two as input and outputs energy their interaction. Only a function that accounts interatomic interactions involved in can accurately predict affinity on unseen molecules. However, make predictions based dataset biases rather than an understanding physics binding. These perform well...
Abstract Despite recent interest in deep generative models for scaffold elaboration, their applicability to fragment-to-lead campaigns has so far been limited. This is primarily due inability account local protein structure or a user’s design hypothesis. We propose novel method fragment STRIFE that overcomes these issues. takes as input Fragment Hotspot Maps (FHMs) extracted from target, and processes them provide meaningful interpretable structural information its model, which turn able...