Computational Chemoproteomics to Understand the Role of Selected Psychoactives in Treating Mental Health Indications

Abstract

<div> <div> <div> <p>We have developed the Computational Analysis of Novel Drug Opportunities (CANDO) platform to infer homology of drug behavior at a proteomic level by constructing and analyzing structural compound-proteome interaction signatures of 3,733 compounds with 48,278 proteins in a shotgun manner. We applied the CANDO platform to predict putative therapeutic properties of 428 psychoactive compounds that belong to phenylethylamine, tryptamine, and cannabinoid chemical classes for treating mental health indications. Our findings indicate that these 428 psychoactives are among the top-ranked predictions for a significant fraction of mental health indications, demonstrating a significant preference for treating such indications over non-mental health indications, relative to randomized controls (p-value < 10-12). Also, we analyzed the use of specific tryptamines for the treatment of sleeping disorders, bupropion for substance abuse disorders, and cannabinoids for epilepsy. Our innovative use of the CANDO platform may guide the identification and development of novel therapies for mental health indications and provide an understanding of their causal basis on a detailed mechanistic level. These predictions can be used to provide new leads for preclinical drug development for mental health and other neurological disorders. </p> </div> </div> </div>

Evaluating Performance of Drug Repurposing Technologies

“…We have used our AIA metric in CANDO extensively [7][8][9][10][11][12][13][14][15][16][17] (section "Drug/compound characterization, benchmarking, evaluation metrics, and performance"). Similarly, Peyvandipour et al used a custom evaluation metric in their systematic drug repurposing study [83].…”

“…We developed and deployed the CANDO platform to model the relationships between every disease/indication and every human use drug/compound [7][8][9][10][11][12][13][14][15][16][17]. Built upon the premise of polypharmacology and multitargeting, at the core of CANDO is the ability to infer similarity of compound/drug behavior.…”

“…Since the development and application of CANDO version 1 [7][8][9][10], we have continued to enhance our platform by analyzing the effect of protein subsets on drug behavior, implementing heterogeneous measures of drug/compound similarity, using multiple molecular docking software packages to evaluate interactions, and refining non-similarity based approaches for drug repurposing in situations where there there is no approved drug for a disease/indication [11][12][13][14][15][16][17]38].…”

“…Less defined end goals (high risk) enable researchers to systematically disrupt the entire drug discovery and development process (high reward). Drug repurposing technologies such as our Computational Analysis of Novel Drug Opportunities (CANDO) platform [7][8][9][10][11][12][13][14][15][16][17] may be used to systematically predict the relative efficacy of every drug in its comprehensive library to treat every disease/indication, minimizing risks and amplifying rewards. In conjunction with mechanistic basic science analyses, these platforms may be used to better understand the science of drug behavior and thereby model reality with greater fidelity.…”

Combined Molecular Graph Neural Network and Structural Docking Selects Potent Programmable Cell Death Protein 1/Programmable Death-Ligand 1 (PD-1/PD-L1) Small Molecule Inhibitors

“…29 Recently, it has been shown that molecular sub-graph features incorporated through a GNN and protein features encoded by their sequence can be combined to predict if a compound can target a given protein. 24 Inspired by this work and based on our interest in developing methods for drug design and immunology [29][30][31][32][33][34][35][36] , we have developed a new machine learning model to predict if a compound can inhibit the PD-1/PD-L1 interaction. Our method replaces the protein sequence features with docking scores representing the free energy of binding and due to this global energetic interaction of the small molecule in the binding pocket, we have termed this model as an "Energy Graph Neural Network" (EGNN).…”