Dataset for "Machine Learning–Guided Design of Mechanoadaptive Bioglues for Multi-Tissue Trauma and First-Aid Applications"

This dataset supports the research on machine learning (ML)-guided rational design of mechanoadaptive bioglues (TuneGlues) for multi-tissue trauma repair and emergency first-aid applications. It contains the full raw experimental data generated from the study, including:Chemical synthesis and structural characterization data of 4 polyurethane prepolymers and 2 amino acid-mimetic crosslinkers (the building blocks of TuneGlues);Mechanical property testing data of 72 groups of TuneGlues formulations, which serves as the training dataset for ML model construction;Raw data of ML model training, 5-fold cross-validation, performance evaluation, and the 100,000-group formulation-mechanical property prediction database;In vitro and in vivo experimental data of TuneGlues, including tissue adhesion performance, biocompatibility, degradation kinetics, hemostatic efficacy, and long-term adhesion stability;Preclinical validation data of TuneGlues in porcine and goat models of single-tissue (lung, intestine, skin, bone) and multi-tissue trauma, as well as performance verification data of the integrated portable first-aid delivery device.This dataset can be reused for the development of novel tissue adhesives, ML-driven biomaterial design, wound healing research, and emergency medical device development. The data correspond to the peer-reviewed and accepted manuscript entitled Machine Learning–Guided Design of Mechanoadaptive Bioglues for Multi-Tissue Trauma and First-Aid Applications.