2020/10/07 by Tristan Bereau, Bereau, Tristan
Biochemistry, Genetics and Molecular Biology · Engineering · Materials Science · #Chemical Physics (physics.chem-ph) #FOS: Physical sciences #Machine Learning in Materials Science #Protein Structure and Dynamics #Soft Condensed Matter (cond-mat.soft) #Surface Chemistry and Catalysis
paper · pdf · doi:10.48550/arxiv.2010.03298
openalex publication_date 2020/10/07 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
Decades of hardware, methodological, and algorithmic development have\npropelled molecular dynamics (MD) simulations to the forefront of\nmaterials-modeling techniques, bridging the gap between electronic-structure\ntheory and continuum methods. The physics-based approach makes MD appropriate\nto study emergent phenomena, but simultaneously incurs significant\ncomputational investment. This topical review explores the use of MD outside\nthe scope of individual systems, but rather considering many compounds. Such an\nin silico screening approach makes MD amenable to establishing coveted\nstructure--property relationships. We specifically focus on biomolecules and\nsoft materials, characterized by the significant role of entropic contributions\nand heterogeneous systems and scales. An account of the state of the art for\nthe implementation of an MD-based screening paradigm is described, including\nautomated force-field parametrization, system preparation, and efficient\nsampling across both conformation and composition. Emphasis is placed on\nmachine-learning methods to enable MD-based screening. The resulting framework\nenables the generation of compound--property databases and the use of advanced\nstatistical modeling to gather insight. The review further summarizes a number\nof relevant applications.\n