This invention proposes a high-
throughput screening method and
system for thermodynamically stable molecular
crystal forms, belonging to the field of molecular
crystal technology, to solve the problems of inefficiency and inaccuracy in existing methods for screening the most thermodynamically stable
crystal forms. The method includes: Step S1, calculating the lattice energy of the molecular crystal using a
quantum mechanical method considering dispersion interactions, and establishing a lattice energy
database containing molecular crystal structures and their corresponding lattice energies; Step S2, based on the lattice energy
database, constructing a quantitative prediction model between
crystal structure and lattice energy using a deep graph neural network, wherein the quantitative prediction model is used to predict the lattice energy corresponding to an unknown
crystal structure; Step S3,
coupling the trained quantitative prediction model with the
crystal structure prediction process; generating candidate crystal structures of the target molecular crystal using the crystal structure prediction method, using the quantitative prediction model to predict the lattice energy of the candidate crystal structures, and screening out thermodynamically stable candidate crystal forms based on the predicted lattice energy.