SYSTEMS AND METHODS FOR ARTIFICIAL INTELLIGENCE-BASED PREDICTION OF AMINO ACID SEQUENCES AT A BINDING INTERFACE
Presented herein are systems and methods for prediction of protein interfaces for binding to target molecules. In certain embodiments, technologies described herein utilize graph-based neural networks to predict portions of protein/peptide structures that are located at an interface of custom biologic (e.g., a protein and/or peptide) that is being designed for binding to a target molecule, such as another protein or peptide. In certain embodiments, graph-based neural network models described herein may receive, as input, a representation (e.g., a graph representation) of a complex comprising a target and a partially-defined custom biologic. Portions of the partially-defined custom biologic may be known, while other portions, such an amino acid sequence and/or particular amino acid types at certain locations of an interface, are unknown and/or to be customized for binding to a particular target. A graph-based neural network model as described herein may then, based on the received input, generate predictions of likely acid sequences and/or types of particular amino acids at the unknown portions. These predictions can then be used to determine (e.g., fill in) amino acid sequences and/or structures to complete the custom biologic.
1 - 28 . (canceled)
29 . A method for the in-silico design of an amino acid interface of a biologic for binding to a target, the method comprising:
(a) receiving, by a processor of a computing device, an initial scaffold-target complex graph comprising a graph representation of at least a portion of a biologic complex comprising the target and a peptide backbone of the biologic;
(b) generating, by the processor, using a machine learning model, a predicted interface comprising, for each of a plurality of interface sites, an identification of a particular amino acid side chain type; and
(c) providing the predicted interface for use in designing the amino acid interface of biologic and/or using the predicted interface to design the amino acid interface of biologic.
30 . A system for the in-silico design of an amino acid interface of a biologic for binding to a target, the system comprising:
a processor of a computing device; and
a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to:
(a) receive an initial scaffold-target complex graph comprising a graph representation of at least a portion of a biologic complex comprising the target and a peptide backbone of the biologic;
(b) generate, using a machine learning model, a predicted interface comprising, for each of a plurality of interface sites, an identification of a particular amino acid side chain type; and
(c) provide the predicted interface for use in designing the amino acid interface of biologic and/or use the predicted interface to design the amino acid interface of the biologic.
31 . The method of claim 29 , wherein the initial scaffold-target complex graph comprises a plurality of nodes and edges.
32 . The method of claim 29 , wherein the initial scaffold-target complex graph comprises a scaffold graph representing at least a portion of the peptide backbone of the biologic, the scaffold graph comprising a plurality of scaffold nodes, each representing a particular amino acid site of the peptide backbone.
33 . The method of claim 32 , wherein a subset of the scaffold nodes are unknown interface nodes, each representing a particular amino acid interface site located in proximity to the target and having an unknown, to-be-determined amino acid side chain.
34 . The method of claim 32 , wherein a subset of the scaffold nodes are known scaffold nodes, each representing a particular amino acid site having a known side chain type.
35 . The method of claim 32 , wherein the scaffold graph comprises a plurality of scaffold edges, each associated with two particular scaffold nodes and representing a relative position and/or orientation of two amino acid sites represented by the two particular scaffold nodes.
36 . The method of claim 29 , wherein the target is or comprises a protein and/or a peptide and the initial scaffold-target complex graph comprises a target graph comprising a plurality of target nodes, each representing a particular amino acid site of the target.
37 . The method of claim 36 , wherein the target graph comprises a plurality of target edges, each associated with two particular target nodes and representing a relative position and/or orientation of two amino acid sites represented by the two particular target nodes.
38 . The method of claim 29 , wherein the machine learning model is or comprises a graph neural network.
39 . The system of claim 30 , wherein the initial scaffold-target complex graph comprises a plurality of nodes and edges.
40 . The system of claim 30 , wherein the initial scaffold-target complex graph comprises a scaffold graph representing at least a portion of the peptide backbone of the biologic, the scaffold graph comprising a plurality of scaffold nodes, each representing a particular amino acid site of the peptide backbone.
41 . The system of claim 40 , wherein a subset of the scaffold nodes are unknown interface nodes, each representing a particular amino acid interface site located in proximity to the target and having an unknown, to-be-determined amino acid side chain.
42 . The system of claim 40 , wherein a subset of the scaffold nodes are known scaffold nodes, each representing a particular amino acid site having a known side chain type.
43 . The system of claim 40 , wherein the scaffold graph comprises a plurality of scaffold edges, each associated with two particular scaffold nodes and representing a relative position and/or orientation of two amino acid sites represented by the two particular scaffold nodes.
44 . The system of claim 30 , wherein the target is or comprises a protein and/or a peptide and the initial scaffold-target complex graph comprises a target graph comprising a plurality of target nodes, each representing a particular amino acid site of the target.
45 . The system of claim 44 , wherein the target graph comprises a plurality of target edges, each associated with two particular target nodes and representing a relative position and/or orientation of two amino acid sites represented by the two particular target nodes.
46 . The system of claim 30 , wherein the machine learning model is or comprises a graph neural network.