Related Experiment Video
Updated: May 13, 2026
Robust Ligature-Induced Model of Murine Periodontitis for the Evaluation of Oral Neutrophils
Published on: January 21, 2020
Machine learning-driven discovery of therapeutic nucleoside hydrogels for periodontitis
Weiqi Li1,2, Yinghui Wen1, Zhenyuan Huang1
1State Key Laboratory of Oral Diseases & National Center for Stomatology & National Clinical Research Center for Oral Diseases & Research Unit of Oral Carcinogenesis and Management & Chinese Academy of Medical Sciences, West China Hospital of Stomatology, Sichuan University, Chengdu, China.
This study uses machine learning to predict nucleoside bioactivity for designing supramolecular hydrogels. GMP and dGMP hydrogels show promise for treating periodontitis, advancing drug delivery and tissue engineering.
Area of Science:
- Biomaterials Science
- Medicinal Chemistry
- Computational Biology
Background:
- Supramolecular hydrogels offer unique properties for drug delivery and tissue engineering.
- Predicting nucleoside bioactivity is crucial for developing effective hydrogel-based therapeutics.
- Rational design of bioactive hydrogels requires robust predictive models.
Purpose of the Study:
- To predict the biological activity of nucleosides for guiding hydrogel synthesis.
- To develop and validate machine learning models for assessing nucleoside bioactivity.
- To identify and test novel nucleoside-based hydrogels for biomedical applications.
Main Methods:
- Feature-selected machine learning models (decision trees, logistic regression, random forest, extreme gradient boosting) were employed.
- Nine predictive models and databases for biological activities were constructed.
- The Molecular Bioactivity Specificity Index (MBSI) and Composite Molecular Attribute Score (CMAS) were introduced for evaluation.
Main Results:
- Two candidate hydrogels, GMP and dGMP, were identified with excellent hydrogel-forming ability, biocompatibility, and antibacterial activity.
- Screening strategies for bioactive nucleoside hydrogels were successfully established.
- The identified GMP and dGMP hydrogels demonstrated efficacy in treating periodontitis models.
Conclusions:
- Machine learning-based strategies, MBSI, and CMAS are effective for rationally designing bioactive nucleoside hydrogels.
- GMP and dGMP hydrogels represent promising candidates for biomedical applications, particularly in targeted therapies for oral diseases.
- This approach facilitates the development of advanced materials for drug delivery and tissue engineering.