A Deep Learning-Based Approach for Identifying the Medicinal Uses of Plant-Derived Natural Compounds

Abstract

Medicinal plants and their extracts have been used as important sources for drug discovery. In particular, plant-derived natural compounds, including phytochemicals, antioxidants, vitamins, and minerals, are gaining attention as they promote health and prevent disease. Although several in vitro methods have been developed to confirm the biological activities of natural compounds, there is still considerable room to reduce time and cost. To overcome these limitations, several in silico methods have been proposed for conducting large-scale analysis, but they are still limited in terms of dealing with incomplete and heterogeneous natural compound data. Here, we propose a deep learning-based approach to identify the medicinal uses of natural compounds by exploiting massive and heterogeneous drug and natural compound data. The rationale behind this approach is that deep learning can effectively utilize heterogeneous features to alleviate incomplete information. Based on latent knowledge, molecular interactions, and chemical property features, we generated 686 dimensional features for 4,507 natural compounds and 2,882 approved and investigational drugs. The deep learning model was trained using the generated features and verified drug indication information. When the features of natural compounds were applied as input to the trained model, potential efficacies were successfully predicted with high accuracy, sensitivity, and specificity.

EGFR and ERK activation resists flavonoid quercetin‑induced anticancer activities in human cervical cancer cells in vitro

“…At present, the morbidity and mortality rates of cervical cancer are at the forefront of malignant tumors, which seriously threatens the health of women (16). In recent years, plant-derived compounds, with the potential to inhibit the development and the proliferation of tumors, have attracted increased attention from scientific researchers (17)(18)(19). In the present study, based on network pharmacology, Que is one of the effective ingredients of SH, then the hub gene, EGFR was identified, which could be associated with Que.…”

Systems pharmacology approaches in herbal medicine research: a brief review

“…However, new therapeutic opportunities for numerous herbal compounds are yet to be identified. In this section, we will review a few studies applying state-of-the-art machine learning methods to prioritize therapeutically effective herbal compounds for several diseases ( 55 - 58 ).…”

A Brief Review of Machine Learning-Based Bioactive Compound Research

“…Another study suggests caffeine as an anti-diabetic NP through an iterative stochastic elimination (ISE) model built on in-house data and confirmation with wet experiments [79]. Recently, a deep learning based model using latent knowledge, molecule to target interactions, and compound property as heterogeneous input features predicted several NPs or NP-derivatives for various cardiovascular diseases (CVD): ergosterol and arginine for heart failure; reserpine, norepinephrine, octopamine, and digitoxin for hypertension; resveratrol for myocardial infarction; and aspirin and agmatine for stroke [80].…”