Prediction of herbal compatibility for colorectal adenoma treatment based on graph neural networks DOI Creative Commons
Li-Mei Gu,

Yinuo Ma,

Shaopeng Liu

et al.

Chinese Medicine, Journal Year: 2025, Volume and Issue: 20(1)

Published: March 5, 2025

Colorectal adenoma is a common precancerous lesion with high risk of malignant transformation. Traditional Chinese medicine and its complex prescriptions have shown promising efficacy in the treatment adenomas; however, there remains lack systematic understanding regarding compatibility patterns within these prescriptions, as well an effective model for predicting therapeutic outcomes. In this study, we collected numerous TCM their components, recommended by experts colorectal adenoma, developed heterogeneous graph neural network to predict strength probability among herbs prescriptions. This delineates relationships herbs, active compounds, molecular targets, allowing quantification interactions potential herbs. Using model, identified high-potential from clinical prescription records components through pharmacology. Through approach, aim provide theoretical foundation foster discovery new optimize TCM, ultimately advance field cancer prevention based on traditional medicine.

Language: Английский

Prediction of herbal compatibility for colorectal adenoma treatment based on graph neural networks DOI Creative Commons
Li-Mei Gu,

Yinuo Ma,

Shaopeng Liu

et al.

Chinese Medicine, Journal Year: 2025, Volume and Issue: 20(1)

Published: March 5, 2025

Colorectal adenoma is a common precancerous lesion with high risk of malignant transformation. Traditional Chinese medicine and its complex prescriptions have shown promising efficacy in the treatment adenomas; however, there remains lack systematic understanding regarding compatibility patterns within these prescriptions, as well an effective model for predicting therapeutic outcomes. In this study, we collected numerous TCM their components, recommended by experts colorectal adenoma, developed heterogeneous graph neural network to predict strength probability among herbs prescriptions. This delineates relationships herbs, active compounds, molecular targets, allowing quantification interactions potential herbs. Using model, identified high-potential from clinical prescription records components through pharmacology. Through approach, aim provide theoretical foundation foster discovery new optimize TCM, ultimately advance field cancer prevention based on traditional medicine.

Language: Английский

Citations

0