Predicting Meridian in Chinese Traditional Medicine Using Machine Learning Approaches

Wang, Y.; Jafari, M.; Tang, Y.; Tang, J.

doi:10.1101/697037

Home/Archive/bioRxiv

bioRxiv · 10.1101/697037

Predicting Meridian in Chinese Traditional Medicine Using Machine Learning Approaches

Wang, Y.·Jafari, M.·Tang, Y.·Tang, J.

View original source ↗Export metadata ↓

Abstract

Plant-derived nature products, known as herb formulas, have been commonly used in Traditional Chinese Medicine (TCM) for disease prevention and treatment. The herbs have been traditionally classified into different categories according to the TCM Organ systems known as Meridians. Despite the increasing knowledge on the active components of the herbs, the rationale of Meridian classification remains poorly understood. In this study, we took a machine learning approach to explore the classification of Meridian. We determined the molecule features for 646 herbs and their active components including structure-based fingerprints and ADME properties (absorption, distribution, metabolism and excretion), and found that the Meridian can be predicted by machine learning approaches with a top accuracy of 0.83. We also identified the top compound features that were important for the Meridian prediction. To the best of our knowledge, this is the first time that molecular properties of the herb compounds are associated with the TCM Meridians. Taken together, the machine learning approach may provide novel insights for the understanding of molecular evidence of Meridians in TCM.\n\nAuthor SummaryIn East Asia, plant-derived natural products, known as herb formulas, have been commonly used as Traditional Chinese Medicine (TCM) for disease prevention and treatment. According to the theory of TCM, herbs can be classified as different Meridians according to the balance of Yin and Yang, which are commonly understood as metaphysical concepts. Therefore, the scientific rational of Meridian classification remains poorly understood. The aim of our study was to provide a computational means to understand the classification of Meridians. We showed that the Meridians of herbs can be predicted by the molecular and chemical features of the ingredient compounds, suggesting that the Meridians indeed are associated with the properties of the compounds. Our work provided a novel chemoinformatics approach which may lead to a more systematic strategy to identify the mechanisms of action and active compounds for TCM herbs.

Source connections

Published article ↗

Explore related subjects

pharmacology and toxicology Preprint

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Wang, Y., Jafari, M., Tang, Y., Tang, J.. 2019-07-09. Predicting Meridian in Chinese Traditional Medicine Using Machine Learning Approaches. https://doi.org/10.1101/697037

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

PFOS and GenX Induce Mitochondrial Toxicity and Differentially Alter Respiratory Complex Activities in Liver and Brain

Per- and polyfluoroalkyl substances (PFAS) are persistent environmental contaminants associated with adverse health effects, yet the mitochondrial consequences of exposure to individual PFAS and their mixtures remain incompletely understood. This study investigated the individual and combined effects of perfluorooctanesulfonic acid (PFOS) and hexafluoropropylene oxide dimer acid (GenX) on mitochondrial respiratory complex activities under acute, sub-chronic, and chronic exposure conditions. Acute effects were evaluated using isolated mouse liver and brain mitochondria, while repeated exposure was assessed in mice treated with PFOS, GenX, or their combination at 0.2, 2, and 10 mg/kg body weight for 1 month or 0.2 mg/kg for 5 months. PFAS distribution in plasma and brain tissues, mitochondrial oxidative phosphorylation (OXPHOS) protein expression, and predicted interactions with mitochondrial Complexes III and IV were also examined. Acute exposure inhibited Complex III and IV activities, with combined treatment generally producing greater inhibition than either compound alone. Following sub-chronic exposure, mitochondrial responses varied according to dose, treatment, and tissue, with more pronounced reductions in liver complex activities and comparatively limited changes in brain enzyme activities. Chronic low-dose exposure altered selected brain mitochondrial complex activities, while PFAS measurements revealed differences in plasma concentration and distribution across brain regions. OXPHOS protein expression also varied across brain regions and exposure conditions. Molecular docking indicated distinct predicted interaction patterns, with PFOS generally exhibiting more favorable scores at Complex III binding sites and GenX at selected Complex IV regions. Overall, these findings demonstrate exposure-dependent and tissue-specific mitochondrial responses to PFOS and GenX and highlight the importance of evaluating combined PFAS exposure across multiple exposure durations.

pharmacology and toxicology↗

bioRxiv2026

Target Specific Conformational Flexibility and Electrostatic Complementarity Modulate LTNF10 Peptide Binding to Phospholipase A2 Variants from Snake Venoms

Phospholipase A2 is one of the primary cytotoxins present in snake venoms, causing tissue damage through the hydrolysis of phospholipids in the cellular membrane. We prioritised and investigated in detail the binding and conformational dynamics of a ten-residue short peptide (LTNF10) from Lethal Toxin Neutralizing Factor from Opossum against snake venom Phospholipase A2 (svPLA2) variants. LTNF10 was prioritised from the curated list of peptide inhibitors. A charge-modulating mutation from aspartic acid to lysine at the fifth position in LTNF10 was observed to enhance the binding affinity to svPLA2 variants from Elapids. The mutant LTNF10 shows selectively enhanced binding affinity toward Group I svPLA2 from Elapidae. Free energy landscapes of LTNF10 and mutant LTNF10, and binding pocket-volume analysis of svPLA2 revealed that Lys5 substitution increases conformational flexibility of mutant LTNF10, allowing it to form favourable binding conformations, while svPLA2 shows a reduction in pocket volume. Conversely, the mutant LTNF10 shows reduced binding affinity in Group II svPLA2 from Viperidae while demonstrating reduced flexibility and a less significant decrease in binding pocket volume in svPLA2. These findings demonstrate that active-site charge microenvironments, binding pocket and structural differences in catalytically relevant loop regions influence LTNF10 binding to svPLA2. The differential binding affinity was primarily driven by improved non-covalent interactions, specifically reduced electrostatic repulsions and enhanced electrostatic interactions in Group I svPLA2 from Elapids. This was influenced by the enhanced flexibility of mutant LTNF10, suggesting the role of a complex scenario of enthalpy-entropy compensation. The understanding of dynamic and structural mechanisms of LTNF10 and svPLA2 interactions and binding offers key insights for designing target-specific peptide therapeutics against diverse svPLA2 toxins.

pharmacology and toxicology↗

bioRxiv2026

Long-PDE4D degradation attenuates left ventricular cardiomyocyte hypertrophic remodelling

Spatial compartmentalisation enables cyclic AMP (cAMP) to regulate distinct cellular functions, yet selectively targeting disease-associated signalling remains challenging. Phosphodiesterases constrain local cAMP responses, but conventional inhibitors offer limited isoform selectivity. Here we investigate whether targeted degradation of long phosphodiesterase-4D (PDE4D) isoforms can suppress pathological growth while preserving contractile function. Using adult rat and human ventricular cardiomyocytes and human myocardial slices, we show that the proteolysis-targeting chimera KTX-184 enhances compartmentalised cAMP signalling and attenuates hypertrophic remodelling. In rat cardiomyocytes, KTX-184 increases {beta}-adrenergic cAMP responses in the cytosol and at the plasma membrane, including T-tubule and crest regions, and enhances nuclear protein kinase A activity. In contrast, phospholamban- and ryanodine receptor-associated cAMP nanodomains, calcium transients and sarcomere shortening remain unchanged. KTX-184 prevents angiotensin II-induced hypertrophic remodelling, preserves T-tubule organisation and normalises hypertrophic marker expression. In human cardiomyocytes, it increases cytosolic cAMP responses and prevents overload-induced hypertrophy in myocardial slices. These findings identify long PDE4D isoforms as targets for spatially selective modulation of cAMP signalling and support targeted degradation as an approach to suppress cardiac hypertrophy without increasing contractility.

pharmacology and toxicology↗