Methods, systems and apparatus, including computer programs encoded on computer storage media. One of the methods includes obtaining a plurality of images of a macromolecule having a plurality of atoms, training a decoder neural network on the plurality of images, and after the training, generating a plurality of conformations for at least a portion of the macromolecule that each include respective three-dimensional coordinates of each of the plurality of atoms, wherein generating each conformation includes sampling a conformation latent representation from a prior distribution over conformation latent representations, processing a respective input including the sampled conformation latent representation using the decoder neural network to generate a conformation output that specifies three-dimensional coordinates of each of the plurality of atoms for the conformation, and generating the conformation from the conformation output.

Methods, systems and apparatus, including computer programs encoded on computer storage media. One of the methods includes obtaining a plurality of images of a macromolecule having a plurality of atoms, training a decoder neural network on the plurality of images, and after the training, generating a plurality of conformations for at least a portion of the macromolecule that each include respective three-dimensional coordinates of each of the plurality of atoms, wherein generating each conformation includes sampling a conformation latent representation from a prior distribution over conformation latent representations, processing a respective input including the sampled conformation latent representation using the decoder neural network to generate a conformation output that specifies three-dimensional coordinates of each of the plurality of atoms for the conformation, and generating the conformation from the conformation output.

A computer-implemented method for predicting a conformation of a ligand docked into a protein is disclosed. According to some embodiments, the method may include determining one or more poses of the ligand in the protein, the poses being representative conformations of the ligand. The method may also include determining, using a neural network, energy scores of the poses. The method may further include determining a proper conformation for the docked ligand based on the energy scores.

A computer-implemented method for predicting a conformation of a ligand docked into a protein is disclosed. According to some embodiments, the method may include determining one or more poses of the ligand in the protein, the poses being representative conformations of the ligand. The method may also include determining, using a neural network, energy scores of the poses. The method may further include determining a proper conformation for the docked ligand based on the energy scores.

This disclosure relates to application based design of novel materials. Conventional methods utilize laborious experimentation or costly first principles calculations. Conventional data driven techniques use point cloud-based representation for crystal structures, that suffers from permutation variance which is not inbuilt in a material's representation, the DL model has to learn invariance which may be inaccurate. Other methods use image based representation for crystal structures and separate images for each element type to represent the basis, which is memory and time intensive. Since each element is represented by its own image, it is difficult for model to learn chemical environment and neighborhood pattern of each element. The embodiments used image based representation of materials consistent with physical principles. Also, embodiments utilize elements matrix to obtain atoms and their positions from basis images. Thus, any material, irrespective of lattice geometry, and number and types of elements, is represented by only two images.

Described herein are systems, methods, and apparatus for electronically drawing and editing representations of chemical structures using an intuitive user interface. This user interface, the context-aware virtual keyboard, makes it faster and easier to draw and edit chemical structure representations by guiding the user through the sequence of steps required to produce the representation in a context-based, non-linear fashion. The context-based virtual keyboard allows a user to quickly create graphical representations of complex chemical structures by using the structure itself as a basis for presenting efficient options for subsequent drawing/editing steps. Different possible and/or likely actions (e.g., edits to a chemical structure being drawn) are presented to the user based on a selected navigation position on the drawing. Thus, a user can efficiently and intuitively modify a chemical structure drawing without the tedious manual selection of portions of the chemical structure and without searching through complicated menus.

Described herein are systems, methods, and apparatus for electronically drawing and editing representations of chemical structures using an intuitive user interface. This user interface, the context-aware virtual keyboard, makes it faster and easier to draw and edit chemical structure representations by guiding the user through the sequence of steps required to produce the representation in a context-based, non-linear fashion. The context-based virtual keyboard allows a user to quickly create graphical representations of complex chemical structures by using the structure itself as a basis for presenting efficient options for subsequent drawing/editing steps. Different possible and/or likely actions (e.g., edits to a chemical structure being drawn) are presented to the user based on a selected navigation position on the drawing. Thus, a user can efficiently and intuitively modify a chemical structure drawing without the tedious manual selection of portions of the chemical structure and without searching through complicated menus.

A system and method for extrapolating a set of specific representational identifiers that are represented or covered by a generic representational identifier found in a target document. Queries are constructed and performed on a corpus of source documents in which members of the extrapolated set of specific representational identifiers are compared to a database of representational data. By matching representational data in this way, any overlap between the generic representational data and specific instances of the generic representational identifier within the source documents is determined. In a more specific implementation, the system and method reduces the scope of the generic representational identifier such that the reduced scope generic representational identifier encompasses only novel specific representational identifiers.

An objective of the present invention is to provide a document creation assistance server and document creation assistance method which are capable of efficiently acquiring information necessary to write a specification or other such documents, and of generating a document from the acquired information. According to an embodiment, this document creation assistance server for assisting with the creation of a document comprises: a final product identification part for accepting identification information for identifying a final product generated by synthesizing a plurality of chemical compounds; a chemical equation acquisition part for referring to electronic experiment note data which provides an electronic record of the details of an experiment relating to the synthesis of the chemical compounds, and acquiring a chemical equation for generating the final product associated with the accepted identification information; and a document generation part for generating a document including the acquired chemical equation.

An objective of the present invention is to provide a document creation assistance server and document creation assistance method which are capable of efficiently acquiring information necessary to write a specification or other such documents, and of generating a document from the acquired information. According to an embodiment, this document creation assistance server for assisting with the creation of a document comprises: a final product identification part for accepting identification information for identifying a final product generated by synthesizing a plurality of chemical compounds; a chemical equation acquisition part for referring to electronic experiment note data which provides an electronic record of the details of an experiment relating to the synthesis of the chemical compounds, and acquiring a chemical equation for generating the final product associated with the accepted identification information; and a document generation part for generating a document including the acquired chemical equation.