Machine learning framework to predict pharmacokinetic profile of small molecule drugs based on chemical structure

Abstract

Accurate prediction of a new compound's pharmacokinetic (PK) profile is pivotal for the success of drug discovery programs. An initial assessment of PK in preclinical species and humans is typically performed through allometric scaling and mathematical modeling. These methods use parameters estimated from in vitro or in vivo experiments, which although helpful for an initial estimation, require extensive animal experiments. Furthermore, mathematical models are limited by the mechanistic underpinning of the drugs' absorption, distribution, metabolism, and elimination (ADME) which are largely unknown in the early stages of drug discovery. In this work, we propose a novel methodology in which concentration versus time profile of small molecules in rats is directly predicted by machine learning (ML) using structure-driven molecular properties as input and thus mitigating the need for animal experimentation. The proposed framework initially predicts ADME properties based on molecular structure and then uses them as input to a ML model to predict the PK profile. For the compounds tested, our results demonstrate that PK profiles can be adequately predicted using the proposed algorithm, especially for compounds with Tanimoto score greater than 0.5, the average mean absolute percentage error between predicted PK profile and observed PK profile data was found to be less than 150%. The suggested framework aims to facilitate PK predictions and thus support molecular screening and design earlier in the drug discovery process.

FIGURE 1

Modeling framework utilized in this study. (1) Chemical structures represented as SMILES string were used as input. (2) RDKit package was utilized to extract Fingerprints and Descriptors based on the SMILES string. (3) Feature selection performed to identify key features. (4) Chemical structures were then used as an input to a machine learning model to predict PK parameters identified in step 3, which were then used as an input to a second machine learning model to predict the PK profile.

FIGURE 2

Visualization of the chemical space of compounds used to train the model and the compounds which were used in the test set. T distributed stochastic neighbor embedding (t‐SNE) approach is used to perform dimensionality reduction to help facilitate visualization of chemical space.

FIGURE 3

Scatter plot of concentrations predicted using machine learning framework (y‐axis) versus observed concentrations (x‐axis) for the test dataset.

FIGURE 4

PK profiles (concentration vs. time) predicted using ML framework (red line), PBPK modeling (Green region) overlayed over observed data (Blue dots).

FIGURE 5

Distribution of ratio of AUC_24h observed versus AUC_24h predicted, and ratio of C _max observed versus C _max predicted for predictions generated using ML framework.