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-11436246-B2
Publication Date
2022-09-06
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 discloses a method and system for presenting, on a computing device, a graphical user interface (GUI) of a therapeutic tool that enables users to visualize and interact with a protein design space for a specific application. This design space comprises a plurality of protein sequences, each associated with a respective plurality of biomedical or biochemical activities. The protein sequences are generated by a machine learning model that uses causal inference to execute alternative scenarios, filter a superset of sequences, and define the sequences within the design space.
A key feature of the invention is the two-screen GUI workflow. The first screen presents the entire design space, while users can select one or more query parameters via graphical elements. Based on these parameters, the system generates and presents, in a second screen, a solution space as a subset of the protein sequences that satisfy the specified criteria. Users can select from within these sequences, particularly those not previously generated by the model, to view additional detailed information including candidate drug compounds, interactions, activities, drugs, genes, and pathways.
The background identifies the problem that conventional drug discovery and candidate selection methods are insufficiently efficient and limited due to small or constrained design spaces, making it difficult to identify effective therapeutics for certain diseases and highly complex therapeutic targets. This invention addresses these problems by enabling efficient enlargement of the design space, advanced filtering and visualization, and providing more comprehensive, informative, and interactive selection of candidate protein sequences and their associated biological activities.
Claims Coverage
There are several inventive features detailed in the independent claims, each pertaining to presenting, filtering, and visualizing protein design and solution spaces using machine learning and interactive GUIs for therapeutic applications.
Presenting a design space in a GUI populated by machine learning with causal inference
A method for presenting, on a computing device, a GUI of a therapeutic tool, where the first screen of the GUI displays a design space for a protein for a specific application. The design space contains a plurality of protein sequences, each associated with a respective plurality of biomedical and/or biochemical activities. The plurality of protein sequences is generated by a machine learning model that utilizes causal inference to execute at least one alternative scenario, filtering a superset of protein sequences to generate those included in the design space.
Selecting and filtering the design space using query parameters to generate a solution space
The receipt, via a graphical element in the first screen, of a selection of one or more query parameters of the design space. Based on these query parameters, a second screen of the GUI is presented, displaying a solution space comprised of a subset of the protein sequences, each retaining its set of activities and selected according to the input query parameters.
Interactive selection of ungenerated protein sequences and presentation of detailed information
Enabling users, via a graphical element of the second screen in the GUI, to select a protein sequence from the subset based on an indication that the sequence has not previously been generated by the machine learning model. Upon selection, the second screen presents additional information about the protein sequence, which may include a candidate drug compound, an interaction, an activity, a drug, a gene, a pathway, or combinations of these.
Visualization of solution space with advanced interactive features and data integration
The GUI includes advanced visualization features, such as presenting the solution space as topographical maps, with color-coded clusters representing correlations (energy correlations) among positions within and between protein sequences. The second screen may include portions that display both these clusters and detailed data describing associated objects, such as drug candidates, activities, interactions, drugs, genes, pathways, physical descriptors, folding properties, wave properties, and stability of modification.
Integration of machine learning metrics and business intelligence with GUI selection
The method allows, via additional GUI interactions, the selection and performance of computational trials using different machine learning models, reporting results that include location within the solution space and relevant model performance metrics (such as memory usage, GPU temperature, and processor usage). The GUI can also receive business intelligence inputs, such as a target product profile, and present new solution space subsets that match these criteria.
In summary, the claims broadly cover the presentation of protein design and solution spaces using advanced machine learning and causal inference approaches, highly interactive GUI selection and visualization, and robust integration of sequence and activity data to enable more efficient and informative therapeutic candidate selection.
Stated Advantages
The invention enables efficient enlargement and exploration of the protein design space, overcoming limitations of conventional drug discovery methods constrained by small or fact-limited design spaces.
The GUI-based workflow allows users to visually navigate, filter, and select protein sequences using interactive query parameters, improving explainability and user experience.
Presentation of solution spaces as topographical maps and color-coded clusters distills complex biochemical information into easily understandable and actionable visual formats.
Integration of causal inference in the machine learning model enhances the filtering and generation of protein sequences, leading to superior candidate selection compared to conventional techniques.
The system supports presentation and selection of previously ungenerated protein sequences, enabling discovery of novel drug candidates.
Performance metrics and business intelligence integration promote resource-efficient model selection and alignment with manufacturing or clinical goals.
Documented Applications
Presentation and selection of candidate drug compounds for therapeutic applications, including but not limited to anti-infective, anti-cancer, anti-microbial, anti-viral, anti-fungal, anti-inflammatory, anti-cholinergic, anti-dopaminergic, anti-serotonergic, anti-noradrenergic, anti-prionic, adhesives, sealants, binders, chelates, diagnostic reporters, biopolymers, encapsulation films, flocculants, and desiccants.
Use in peptide therapeutic design tools and peptide business intelligence tools for optimizing sequences based on pharmacology, pharmacokinetics, clinical, and manufacturing data.
Application in GUI-based systems allowing users (such as protein or peptide designers and business users) to identify, analyze, and act upon novel protein sequences tailored to specific biomedical, biochemical, or product profile requirements.
Interested in licensing this patent?