Generating enhanced graphical user interfaces for presentation of anti-infective design spaces for selecting drug candidates
Inventors
Lee, Francis • STECKBECK, Jonathan D. • Holste, Hannes
Assignees
Publication Number
US-11403316-B2
Publication Date
2022-08-02
Expiration Date
2041-05-13
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
In one aspect, a method is disclosed for presenting, on a computing device, a graphical user interface (GUI) of a therapeutic tool. The method includes presenting, in a first screen of the GUI, a design space for a protein for an application, where the design space includes a set of sequences, where each sequence contains a respective set of activities pertaining to the application. The method also includes receiving, via a graphical element in the first screen, a selection of one or more query parameters of the design space, and presenting, in a second screen of the GUI, a solution space that includes a subset of the set of sequences, where each sequence contains the respective set of activities, where the subset of the set of sequences is selected based on the one or more query parameters.
Core Innovation
The invention provides a method and system for presenting, on a computing device, an enhanced graphical user interface (GUI) for a therapeutic tool that enables the visualization and selection of drug candidate design spaces, specifically for proteins and peptides. The GUI presents a design space comprising a plurality of protein sequences, where each sequence is associated with multiple activities relevant to a therapeutic application. These activities include biomedical and biochemical functionalities, such as antimicrobial or anti-cancer properties.
A machine learning model, trained using various encodings and employing causal inference through the simulation of alternative scenarios, generates and filters a superset of protein sequences. Through the GUI, a user can specify one or more query parameters via graphical elements, which filter the design space to generate a solution space presenting a targeted subset of sequences. This subset is visualized in various ways, such as color-coded clusters, topographical maps, or other graphical methods, and differentiated based on whether sequences have been previously generated.
The background describes that conventional drug discovery methods are limited by narrow design spaces and inefficient processes. These approaches often constrain searches to facts or known activities, making it hard to identify therapeutics for challenging or refractory diseases. The patent addresses these limitations by enabling larger, information-rich design spaces, efficient navigation and filtering, and the ability to present complex sequence-activity relationships in a visual and interactive way, facilitating faster and broader exploration and selection of candidate drugs.
Claims Coverage
The claims set forth several inventive features relating to GUI-based selection and visualization of protein design and solution spaces generated by machine learning models using causal inference and encoded data.
Graphical user interface for therapeutic tool presenting protein design spaces
A method for presenting, on a computing device, a GUI of a therapeutic tool that: - Presents, in a first screen, a design space for a protein for an application. The design space contains a plurality of protein sequences, each associated with a plurality of activities pertinent to the application, including biomedical and biochemical activities. - The sequences are generated by a machine learning model trained with multiple encodings and use causal inference to simulate alternative scenarios to filter and generate the protein sequence set.
User-driven selection of query parameters and dynamic generation of solution spaces
Allows the user, via graphical elements in the GUI, to select one or more query parameters affecting the design space. In response, the system presents, in a second GUI screen, a solution space comprising a filtered subset of protein sequences, each retaining their respective plurality of activities, where selection is based on the chosen query parameters.
Visual differentiation of generated and untraversed sequences
Within the solution space displayed in the GUI, the method provides: - Presentation of a first protein sequence in a first manner (e.g., visual distinction), based on a flag indicating that sequence has been previously generated. - Presentation of a second protein sequence in a second manner, based on a flag indicating that it has not been previously generated. This enables users to distinguish between traversed and untraversed sequence candidates.
In summary, the inventive features cover the GUI-based method of presenting and navigating protein design spaces generated by advanced machine learning, the interactive selection and filtering of these spaces by user-defined parameters, and the clear visual differentiation between previously generated and new sequence candidates.
Stated Advantages
Enables efficient enlargement of the design space to include diverse drug sequence, activity, semantic, chemical, and physical information, overcoming the limitations of conventional drug discovery techniques.
Reduces computational complexity and resource consumption by efficiently filtering and presenting candidate sequences, even within very large design spaces.
Facilitates user understanding, exploration, and selection of complex candidate drug spaces via advanced graphical visualizations, such as topographical maps and network clusters.
Supports identification of superior drug candidates with desirable multi-dimensional activity profiles through the integration of machine learning, causal inference, and interactive user input.
Improves explainability and usability for researchers and business users by providing interactive, clear, and easily navigable interfaces that distill large complex datasets into actionable views.
Documented Applications
Anti-infective therapeutic design, including applications for prosthetic joint infections, urinary tract infections, intra-abdominal or peritoneal infections, otitis media, cardiac infections, respiratory infections, neurological infections, dental infections, digestive and intestinal infections, wound and soft tissue infections, and physiological system infections.
Animal health and veterinary applications, such as the treatment of animal diseases (e.g., bovine mastitis).
Industrial applications including anti-biofouling and generating optimized control action sequences for machinery.
Applications for new therapeutic indications such as eczema, inflammatory bowel disease, Crohn's disease, rheumatoid arthritis, asthma, auto-immune diseases, inflammatory disease processes, and oncology treatments and palliatives.
Generating optimized sequences of decisions for non-player character (NPC) behavior in the video game industry.
Integrated circuit/chip design industry, including optimization in mask works generation and routing processes for efficient and high-performance semiconductor devices.
Interested in licensing this patent?