Machine learning driven web-based app platform for the discovery of monoamine oxidase B inhibitors

Abstract

Monoamine oxidases (MAOs), specifically MAO-A and MAO-B, play important roles in the breakdown of monoamine neurotransmitters. Therefore, MAO inhibitors are crucial for treating various neurodegenerative disorders, including Parkinson's disease (PD), Alzheimer's disease (AD), and amyotrophic lateral sclerosis (ALS). In this study, we developed a novel cheminformatics pipeline by generating three diverse molecular feature-based machine learning-assisted quantitative structural activity relationship (ML-QSAR) models concerning MAO-B inhibition. PubChem fingerprints, substructure fingerprints, and one-dimensional (1D) and two-dimensional (2D) molecular descriptors were implemented to unravel the structural insights responsible for decoding the origin of MAO-B inhibition in 249 non-reductant molecules. Based on a random forest ML algorithm, the final PubChem fingerprint, substructure fingerprint, and 1D and 2D molecular descriptor prediction models demonstrated significant robustness, with correlation coefficients of 0.9863, 0.9796, and 0.9852, respectively. The significant features of each predictive model responsible for MAO-B inhibition were extracted using a comprehensive variance importance plot (VIP) and correlation matrix analysis. The final predictive models were further developed as a web application, MAO-B-pred ( https://mao-b-pred.streamlit.app/ ), to allow users to predict the bioactivity of molecules against MAO-B. Molecular docking and dynamics studies were conducted to gain insight into the atomic-level molecular interactions between the ligand-receptor complexes. These findings were compared with the structural features obtained from the ML-QSAR models, which supported the mechanistic understanding of the binding phenomena. The presented models have the potential to serve as tools for identifying crucial molecular characteristics for the rational design of MAO-B target inhibitors, which may be used to develop effective drugs for neurodegenerative disorders.

Keywords: 1D and 2D molecular descriptors; Bioactivity; ML-QSAR; Molecular docking; Molecular dynamics simulation; Molecular interactions; Monoamine oxidase B; Prediction models; PubChem fingerprints; Substructure fingerprints; Web application.

Figure 1

Structures of amide-based MAO inhibitors.

Figure 2

Workflow for the generation of multi-feature ML-QSAR predictive models.

Figure 3

Exploratory data analysis of the curated amide-based MAO-B inhibitor dataset.

Figure 4

Regression plots of all generated ML-QSAR models.

Figure 5

Applicability domain analysis by PCA plot for the generated ML-QSAR models. (A) and (B) show 2D and 3D PCA of the PubChem fingerprint prediction model; (C) and (D) show 2D and 3D PCA of the substructure fingerprint prediction model; (E) and (F) show 2D and 3D analyses of the 1D and 2D molecular descriptor prediction model.

Figure 6

VIP plot analysis of ten optimal features of the PubChem prediction model.

Figure 7

VIP plot analysis of ten optimal features of the substructure prediction model.

Figure 8

VIP plot analysis of ten optimal features of the 1D and 2D molecular descriptor prediction model.

Figure 9

Chemical structures of previously known amide-based MAO-B inhibitors along with newly discovered molecule C175-0062.

Figure 10

Structural analysis of optimal active molecules of the QSAR dataset in contradiction with VIP plot extracted features. (A) ML-QSAR PubChem fingerprint prediction model; (B) ML-QSAR substructure fingerprint prediction model; (C) ML-QSAR 1D 2D molecular descriptor prediction model.

Figure 11

2D (A) and 3D (B) interactions of C175-0062 with the MAO-B binding pocket.

Figure 12

Structural analysis of C175-0062 in contradiction with VIP plot extracted features. (A) ML-QSAR PubChem fingerprint prediction model; (B) ML-QSAR substructure fingerprint prediction model; (C) ML-QSAR 1D 2D molecular descriptor prediction model.

Figure 13

Analysis of the C175-0062 -MAO-B complex using MD simulation. (A) RMSD (protein RMSD is shown in blue, and RMSD of C175-0062 is shown in red). (B) Individual amino acid RMSF for proteins. (C) Analysis of protein–ligand contacts of MD trajectory. (D) 2D Interaction diagram.

Figure 14

PCA of C175-0062-hMAO-B protein–ligand complex.