Comprehensive machine learning boosts structure-based virtual screening for PARP1 inhibitors - Key findings and methods

Aug 4, 2026

Poly ADP-ribose polymerase 1 (PARP1) serves as an important therapeutic target for cancer treatment. Machine-learning scoring functions offer a route to identify novel PARP1 inhibitors, and this study evaluates PARP1-specific models using semi-synthetic docking-activity labeled training data. Harder test sets are built by selecting molecules dissimilar to training compounds. Five supervised learning algorithms with protein–ligand pose fingerprints and ligand-only features identify a single highly predictive scoring function: a PARP1-specific support vector machine regressor using PLEC fingerprints, achieving the strongest top-1% Normalized Enrichment Factor on the hardest set and outperforming classical baselines.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages17

Reading Time43 min

AccessFree to download

Abstract

Key points

Introduction

PARP1 biological role and activation

PARP1 as a cancer therapy target and BRCA-related repair context

Machine-learning scoring functions overview

Dataset construction from docking activity labels

Model evaluation on increasingly difficult test sets

Predictive scoring function and descriptors

Caba etal. Journal ofCheminformatics (2024) 16:40 [https://doi.org/10.1186/s13321-024-00832-1](https://doi.org/10.1186/s13321-024-00832-1) Journal of Cheminformatics RESEARCH Open Access Comprehensive machine learning boosts structure-based virtual screening for PARP1 inhibitors Klaudia Caba1, Viet‑KhoaTran‑Nguyen2, Taufiq Rahman3 and Pedro J. Ballester 1* Abstract Poly ADP‑ribose polymerase 1 (PARP1) is an attractive therapeutic target for cancer treatment. Machine‑learning scor‑ ing functions constitute a promising approach to discovering novel PARP1 inhibitors. Cutting‑edge PARP1‑specific machine‑learning scoring functions were investigated using semi‑synthetic training data from docking activity‑ labelled molecules: known PARP1 inhibitors, hard‑to‑discriminate decoys property‑matched to them with generative graph neural networks and confirmed inactives. We further made test sets harder by including only molecules dissimi‑ lar to those in the training set. Comprehensive analysis of these datasets using five supervised learning algorithms, and protein–ligand fingerprints extracted from docking poses and ligand only features revealed one highly predictive scoring function. This is the PARP1‑specific support vector machine‑based regressor, when employing PLEC finger‑ prints, which achieved a high Normalized Enrichment Factor at the top 1% on the hardest test set (NEF1% = 0 . 588, median of 10 repetitions), and was more predictive than any other investigated scoring function, especially the classi‑ cal scoring function employed as baseline. Key points • A new scoring tool based on machine‑learning was developed to predict PARP1 inhibitors for potential cancer treatment. • The majority of PARP1‑specific machine‑learning models performed better than generic and classical scoring functions. • Augmenting the training set with ligand‑only Morgan fingerprint features generally resulted in better performing models, but not for the best models where no further improvement was observed. • Employing protein‑ligand‑extracted fingerprints as molecular descriptors led to the best‑performing and most‑ efficient model for predicting PARP1 inhibitors. • Deep learning performed poorly on this target in comparison with the simpler ML models. Keywords Structure‑based virtual screening, Machine learning scoring functions, Target‑specific scoring functions, PARP1 inhibitors, Molecular docking *Correspondence: Pedro J. Ballester [p.ballester@imperial.ac.uk](p.ballester@imperial.ac.uk) Full list of author information is available at the end of the article © The Author(s) 2024. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit [http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (](http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver ()[http://creativeco](http://creativeco)[mmons.org/publicdomain/zero/1.0/](mmons.org/publicdomain/zero/1.0/)) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Caba etal. Journal ofCheminformatics (2024) 16:40 Introduction PARP1 plays an important role in regulating the microhomology-mediated

FAQs

What is the main goal of the study on PARP1 inhibitors?

To assess whether PARP1-specific machine-learning scoring functions can improve structure-based virtual screening and better discover novel PARP1 inhibitors for cancer treatment.

How are the training and test datasets constructed?

Training uses semi-synthetic docking activity–labeled molecules including known inhibitors, property-matched decoys from generative graph neural networks, and confirmed inactives; test sets are made harder by including only molecules dissimilar to the training set.

Which scoring model and molecular descriptors performed best?

A PARP1-specific support vector machine-based regressor using PLEC fingerprints achieved the highest top-1% Normalized Enrichment Factor on the hardest test set, outperforming other investigated functions including the classical baseline.

Aug 4, 2026

Upload by Ava Thompson