[Paper Review] Network medicine framework reveals generic herb-symptom effectiveness of Traditional Chinese Medicine

Xiao Gan, Zixin Shu+9 more|arXiv (Cornell University)|Jul 18, 2022

Traditional Chinese Medicine Studies6 citations

TL;DR

This study introduces a network medicine framework that maps herb-symptom relationships in Traditional Chinese Medicine (TCM) using the human protein interactome. By showing that network proximity between a herb's targets and symptom-associated gene modules predicts treatment effectiveness, the framework identifies novel, biologically plausible herb-symptom pairs validated in real hospital data.

ABSTRACT

Traditional Chinese medicine (TCM) relies on natural medical products to treat symptoms and diseases. While clinical data have demonstrated the effectiveness of selected TCM-based treatments, the mechanistic root of how TCM herbs treat diseases remains largely unknown. More importantly, current approaches focus on single herbs or prescriptions, missing the high-level general principles of TCM. To uncover the mechanistic nature of TCM on a system level, in this work we establish a generic network medicine framework for TCM from the human protein interactome. Applying our framework reveals a network pattern between symptoms (diseases) and herbs in TCM. We first observe that genes associated with a symptom are not distributed randomly in the interactome, but cluster into localized modules; furthermore, a short network distance between two symptom modules is indicative of the symptoms' co-occurrence and similarity. Next, we show that the network proximity of a herb's targets to a symptom module is predictive of the herb's effectiveness in treating the symptom. We validate our framework with real-world hospital patient data by showing that (1) shorter network distance between symptoms of inpatients correlates with higher relative risk (co-occurrence), and (2) herb-symptom network proximity is indicative of patients' symptom recovery rate after herbal treatment. Finally, we identified novel herb-symptom pairs in which the herb's effectiveness in treating the symptom is predicted by network and confirmed in hospital data, but previously unknown to the TCM community. These predictions highlight our framework's potential in creating herb discovery or repurposing opportunities. In conclusion, network medicine offers a powerful novel platform to understand the mechanism of traditional medicine and to predict novel herbal treatment against diseases.

Motivation & Objective

• To uncover the systemic, mechanism-based principles underlying TCM herb efficacy beyond single-herb or prescription studies.
• To address the lack of understanding in how TCM herbs treat diseases at the molecular level.
• To develop a generic, system-level framework for predicting herb-symptom effectiveness using network biology.
• To validate predictions using real-world inpatient clinical data on symptom co-occurrence and recovery rates.
• To identify novel, previously unknown herb-symptom pairs with high predictive and clinical potential.

Proposed method

• Construct a human protein-protein interaction (PPI) network to map disease- and herb-associated genes.
• Identify gene modules associated with specific symptoms by clustering disease-related genes in the PPI network.
• Calculate network distance between a herb’s target genes and symptom-associated gene modules as a proxy for therapeutic proximity.
• Use clinical inpatient data to correlate network distances with symptom co-occurrence and recovery rates after herbal treatment.
• Apply statistical and network-based models to test whether shorter network distances predict higher symptom co-occurrence and better recovery.
• Validate predicted novel herb-symptom pairs using real hospital records, confirming their effectiveness beyond known TCM knowledge.

Experimental results

Research questions

• RQ1 Do symptom-associated genes in TCM cluster into localized modules within the human protein interactome?
• RQ2 Is there a correlation between network proximity of a herb’s targets to a symptom’s gene module and the herb’s effectiveness in treating that symptom?
• RQ3 Does a shorter network distance between two symptoms predict their co-occurrence in patients?
• RQ4 Can network-based proximity between herbs and symptoms predict actual patient recovery rates after treatment?
• RQ5 Are there novel herb-symptom pairs predicted by the framework that are confirmed in real clinical data but unknown in traditional TCM literature?

Key findings

• Symptom-associated genes in TCM form localized, biologically coherent modules in the human protein interactome.
• Shorter network distances between a herb’s targets and a symptom’s gene module are significantly correlated with higher treatment effectiveness.
• Inpatient data confirm that symptoms with shorter network distances co-occur more frequently, indicating shared biological mechanisms.
• Herb-symptom network proximity predicts patient recovery rates, with higher proximity linked to faster symptom resolution.
• The framework successfully identified novel herb-symptom pairs not previously recognized in TCM, which were validated in clinical data.
• The network medicine framework provides a predictive platform for herb discovery and repurposing in traditional medicine.