Innovative virtual screening of PD-L1 inhibitors: the synergy of molecular similarity, neural networks and GNINA docking

Future Med Chem. 2024;16(20):2107-2118. doi: 10.1080/17568919.2024.2389773. Epub 2024 Sep 4.

Abstract

Aims: Immune checkpoint inhibitors targeting PD-L1 are crucial in cancer research for preventing cancer cells from evading the immune system.Materials & methods: This study developed a screening model combining ANN, molecular similarity, and GNINA 1.0 docking to target PD-L1. A database of 2044 substances was compiled from patents.Results: For molecular similarity, the AVALON emerged as the most effective fingerprint, demonstrating an AUC-ROC of 0.963. The ANN model outperformed the Random Forest and Support Vector Classifier in cross-validation and external validation, achieving an average precision of 0.851 and an F1 score of 0.790. GNINA 1.0 was validated through redocking and retrospective control, achieving an AUC of 0.975.Conclusions: From 15235 DrugBank compounds, 22 candidates were shortlisted. Among which (3S)-1-(4-acetylphenyl)-5-oxopyrrolidine-3-carboxylic acid emerged as the most promising.

Keywords: GNINA; PD-L1 inhibitors; drug discovery; molecular similarity; virtual screening.

Plain language summary

[Box: see text].

MeSH terms

  • B7-H1 Antigen* / antagonists & inhibitors
  • B7-H1 Antigen* / metabolism
  • Drug Evaluation, Preclinical
  • Humans
  • Immune Checkpoint Inhibitors* / chemistry
  • Immune Checkpoint Inhibitors* / pharmacology
  • Molecular Docking Simulation*
  • Molecular Structure
  • Neural Networks, Computer*

Substances

  • B7-H1 Antigen
  • CD274 protein, human
  • Immune Checkpoint Inhibitors