Abstract: A method for machine learning aided modeling of two interacting structures may include: (a) receiving an input structure comprising an interaction region; (b) generating a plurality of candidate structures using a first differentiable machine learning model; (c) docking one or more candidate structures of the plurality of candidate structures at the interaction region of the input structure using a second differentiable machine learning model to predict a docking geometry; (d) ranking the one or more candidate structures of the plurality of candidate structures docked in (c) using a third differentiable machine learning model to predict a score; and (e) backpropagating the score to (i) the first differentiable machine learning model to update the plurality of candidate structures or (ii) the second differentiable machine learning model to update the docking geometry.

Abstract: Disclosed systems and methods obtain a graph of a polymer comprising nodes and edges, the nodes representing polymer atoms, the edges encoding distances and relative orientations between corresponding pairs of nodes and whether pairs of nodes are covalently bound. Each subgraph in a plurality of first partial-context subgraphs of the graph is sequentially inputted into a first model to calculate first side chain dihedral angles for polymer residues. This updates the graph through first side chain dihedral angles. Each second subgraph in a plurality of second partial-context subgraphs of the updated graph is inputted into a second model, thereby obtaining calculated second side chain dihedral angles for polymer residues that serve to update the graph through second side chain dihedral angles. The graph is again updated with updated side chain dihedral angle values obtained by sequentially inputting full-context subgraphs, each such subgraph representing a different residue, into a full-context model.

Abstract: A method for machine learning aided modeling of two interacting structures may include: (a) receiving an input structure comprising an interaction region; (b) generating a plurality of candidate structures using a first differentiable machine learning model; (c) docking one or more candidate structures of the plurality of candidate structures at the interaction region of the input structure using a second differentiable machine learning model to predict a docking geometry; (d) ranking the one or more candidate structures of the plurality of candidate structures docked in (c) using a third differentiable machine learning model to predict a score; and (e) backpropagating the score to (i) the first differentiable machine learning model to update the plurality of candidate structures or (ii) the second differentiable machine learning model to update the docking geometry.