Assigning Confidence to Molecular Property Prediction

02/23/2021
by   AkshatKumar Nigam, et al.
13

Introduction: Computational modeling has rapidly advanced over the last decades, especially to predict molecular properties for chemistry, material science and drug design. Recently, machine learning techniques have emerged as a powerful and cost-effective strategy to learn from existing datasets and perform predictions on unseen molecules. Accordingly, the explosive rise of data-driven techniques raises an important question: What confidence can be assigned to molecular property predictions and what techniques can be used for that purpose? Areas covered: In this work, we discuss popular strategies for predicting molecular properties relevant to drug design, their corresponding uncertainty sources and methods to quantify uncertainty and confidence. First, our considerations for assessing confidence begin with dataset bias and size, data-driven property prediction and feature design. Next, we discuss property simulation via molecular docking, and free-energy simulations of binding affinity in detail. Lastly, we investigate how these uncertainties propagate to generative models, as they are usually coupled with property predictors. Expert opinion: Computational techniques are paramount to reduce the prohibitive cost and timing of brute-force experimentation when exploring the enormous chemical space. We believe that assessing uncertainty in property prediction models is essential whenever closed-loop drug design campaigns relying on high-throughput virtual screening are deployed. Accordingly, considering sources of uncertainty leads to better-informed experimental validations, more reliable predictions and to more realistic expectations of the entire workflow. Overall, this increases confidence in the predictions and designs and, ultimately, accelerates drug design.

READ FULL TEXT

page 3

page 5

page 8

page 18

research
03/20/2019

Uncertainty quantification of molecular property prediction with Bayesian neural networks

Deep neural networks have outperformed existing machine learning models ...
research
12/15/2020

Molecular machine learning with conformer ensembles

Virtual screening can accelerate drug discovery by identifying top candi...
research
07/19/2023

Uncertainty Quantification for Molecular Property Predictions with Graph Neural Architecture Search

Graph Neural Networks (GNNs) have emerged as a prominent class of data-d...
research
01/03/2018

Rapid, concurrent and adaptive extreme scale binding free energy calculation

The recently demonstrated ability to perform accurate, precise and rapid...
research
08/20/2022

A biologically-inspired evaluation of molecular generative machine learning

While generative models have recently become ubiquitous in many scientif...
research
07/21/2022

Low cost prediction of probability distributions of molecular properties for early virtual screening

While there is a general focus on predictions of values, mathematically ...
research
06/20/2023

MoleCLUEs: Optimizing Molecular Conformers by Minimization of Differentiable Uncertainty

Structure-based models in the molecular sciences can be highly sensitive...

Please sign up or login with your details

Forgot password? Click here to reset