arXiv subjects

Yuuki Ishiwatari

Publications and source records attributed to Yuuki Ishiwatari.

3 recordsLinked to original sources

arXiv2025

Composition-agnostic prediction of self-assembly in multicomponent amphiphile mixtures from molecular structure

Predicting self-assembly in multi-component amphiphilic systems is challenging due to the complexity of intercomponent interactions and the combinatorial growth of possible formulations. In this study, we develop a unified machine-learning framework that directly predicts self-assembly behavior from the molecular structures of constituent components, independent of the number or identity of those components. We extend the critical packing parameter (CPP) to multi-component systems and generate a large dataset of self-assembled morphologies using dissipative particle dynamics (DPD) simulations. By systematically evaluating twelve combinations of feature extraction methods and model architectures, we find that models incorporating a fully connected graph convolutional network (GCN) layer achieve superior performance, with the GCN-GCN architecture accurately capturing both intramolecular relationships and intercomponent interactions. Notably, this model exhibits strong extrapolative capability: it accurately predicts CPP values for five-component mixtures even when trained only on systems with fewer components, and it maintains high accuracy for mixtures composed of molecular species that are entirely absent from the training data. These results demonstrate that a composition-agnostic predictive framework can enable efficient virtual screening and provide a foundation for the rational design of complex amphiphilic materials.

cond-mat.soft↗

arXiv2024

Machine learning-enabled exploration of mesoscale architectures in amphiphilic-molecule self-assembly

Amphiphilic molecules spontaneously form self-assembled structures of various shapes depending on their molecular structures, the temperature, and other physical conditions. The functionalities of these structures are dictated by their formations and their properties must be evaluated for reproduction using molecular simulations. However, the assessment of such intricate structures involves many procedural steps. This study investigates the potential of machine-learning models to extract structural features from mesoscale non-ordered self-assembled structures, and suggests a methodology in which machine-learning models for the structural analysis of self-assembled structures are trained on particle types and coordinate data. In the proposed approach, graph neural networks are utilised to extract local structural data for analysis. In simulations using several hundred self-assembled structures of up to 4050 coarse-grained particles, local structures are successfully extracted and classified with up to 78.35 % accuracy. As the machine-learning models learn structural characteristics without the need for human-made feature engineering, the proposed method has important potential applications in the field of materials science.

cond-mat.soft↗

arXiv2023

Machine learning prediction of self-assembly and analysis of molecular structure dependence on the critical packing parameter

Amphiphilic molecules spontaneously form self-assembly structures based on physical conditions such as molecular structure, concentration, and temperature. These structures exhibit various useful functions according to their morphology. The concept of the critical packing parameter serves to correlate self-organized structures with chemical composition. However, unless both molecular arrangement and self-assembly patterns are understood, direct computational utilization for molecular design remains challenging. In this study, we attempt to predict the self-assembled structure of a molecule directly from its chemical structure and analyze factors influencing it using machine learning. Dissipative particle dynamics simulations were used to reproduce many self-assembly structures composed of various chemical structures, and their critical packing parameters were calculated. A machine learning model was built using the chemical structures as input data and the critical packing parameters as output data.As a result, both Random Forest and a type of Recurrent Neural Network known as GRU demonstrated high predictive accuracy. It has been revealed through feature importance analysis and dependence on sample size that the amphiphilic nature of molecules significantly influences the self-assembly structures. Additionally, the importance of selecting an appropriate molecular structure representation for each algorithm has been emphasized. The results of this research will help to further streamline product development in the fields of materials science, materials chemistry, and medical materials.

cond-mat.mtrl-sci↗