Generating enhanced graphical user interfaces for presentation of anti-infective design spaces for selecting drug candidates
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.
1. A method for presenting, on a computing device, a graphical user interface (GUI) of a therapeutic tool, the method comprising:
presenting, in a first screen of the GUI, a design space for a protein for an application, wherein the design space comprises a plurality of protein sequences, wherein:
each protein sequence is associated with a respective plurality of activities pertaining to the application,
the plurality of activities comprises one or more biomedical activities, biochemical activities, or some combination thereof, and
the plurality of protein sequences is generated by a machine learning model, wherein the machine learning model uses causal inference to execute at least one of a plurality of alternative scenarios to filter a superset of protein sequences and to generate the plurality of protein sequences in the design space;
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 comprising a subset of the plurality of protein sequences, wherein each protein sequence contains the respective plurality of activities, wherein the subset of the plurality of protein sequences is selected based on the one or more query parameters;
receiving, using a graphical element of the second screen, a selection of a protein sequence from the subset of the plurality of protein sequences, wherein the selection is based on an indication that the protein sequence has not been previously generated by the machine learning model; and
responsive to the selection of the protein sequence, presenting, in the second screen, additional information pertaining to the protein sequence, wherein the additional information comprises:
a candidate drug compound, an interaction, an activity, a drug, a gene, a pathway, or some combination thereof.
2. The method of claim 1 , wherein the second screen comprises:
a first portion presenting one or more color-coded clusters representing the subset of the plurality of protein sequences, and
a second portion presenting data pertaining to the subset of the plurality of protein sequences represented by the one or more color-coded clusters, wherein the data describes one or more objects associated with the subset of the plurality of protein sequences, and the one or more objects comprise a candidate drug compound, an activity, an interaction, a drug, a gene, a pathway, a physical descriptor, a characteristic, an interaction, a folding property, a wave property, a stability of modification, or some combination thereof.
3. The method of claim 2 , wherein the one or more color-coded clusters represent, using an energy correlation, each sequence in the subset, and the energy correlation comprises at least one correlation between each position of each protein sequence in the subset and other positions of other protein sequences in the subset.
4. The method of claim 1 , wherein the solution space is presented as a topographical map in the GUI, wherein the topographical map comprises a plurality of indications that each represent a level of activity for a protein sequence associated with a given point on the topographical map.
5. The method of claim 1 , wherein the design space is generated based on a knowledge graph pertaining to peptides and the design space is presented as a two-dimensional (2D) elevation map, a three-dimensional (3D) shape or an n-dimensional (nD) mathematical representation.
6. The method of claim 1 , wherein the solution space is generated within the design space by one or more machine learning models trained to measure, based on the query parameter, a respective level of one or more of the respective plurality of activities of each of the plurality of protein sequences in the subset, wherein the query parameter comprises a sequence parameter.
7. The method of claim 1 , further comprising:
receiving, using a graphical element of the second screen, a selection of a protein sequence from the subset of the plurality of protein sequences; and
presenting, in a third screen of the GUI, a candidate dashboard comprising information pertaining to the protein sequence, wherein the information pertains to a structure of the protein sequence, a correlation heatmap, experimental data, a list of probabilistic scores generated by inference models, external data related to the protein sequence, or some combination thereof.
8. The method of claim 1 , further comprising:
receiving a selection of a trial configured to be performed by a machine learning model, wherein the machine learning model uses the solution space; and
receiving, from an artificial intelligence engine, one or more results of performing the trial, wherein the one or more results:
provide a location of a point reached in the solution space after performing a traversal of the solution space defined by the trial, and
provide a metric of a machine learning model used by the artificial intelligence engine to perform the trial, wherein the metric pertains to one or more of memory usage, graphic processing unit temperature, power usage, processor usage, or central processing unit temperature.
9. The method of claim 1 , further comprising:
receiving, from a graphical element of a business intelligence screen of the GUI, a target product profile, wherein the target product profile comprises pharmacology data, pharmacokinetic data, pharmacodynamic data, activity data, manufacturing data, compliance data, clinical trial data, or some combination thereof;
receiving, from an artificial intelligence engine, a second subset of the plurality of protein sequences, wherein the second subset of the plurality of protein sequences is selected based on the target product profile; and
presenting, in the GUI, the second subset of the plurality of protein sequences.
10. The method of claim 1 , further comprising:
receiving, in the GUI, one or more parameters pertaining to one or more machine learning models of an artificial intelligence engine, wherein the one or more parameters pertain to one or more constraints for the one or more machine learning models to implement when performing one or more trials using the solution space.
11. The method of claim 1 , wherein the therapeutic tool is a peptide therapeutic tool.
12. The method of claim 1 , wherein the protein is a peptide.
13. The method of claim 1 , wherein the one or more query parameters comprise a plurality of biomedically-related ontology terms, a plurality of non-biomedically-related ontology terms, or some combination thereof.
14. The method of claim 13 , wherein the plurality of biomedically-related ontology terms pertains to indications, genes, symptoms, or some combination thereof, and the plurality of non-biomedically-related ontology terms pertain to characteristics, descriptors, or some combination thereof.
15. The method of claim 1 , further comprising:
receiving, using a graphical element of the second screen, a selection of a protein sequence from the subset of the plurality of protein sequences; and
causing the sequence to be analyzed, manufactured, synthesized, or produced.
16. A tangible, non-transitory computer-readable medium storing instructions that, when executed, cause a processing device to:
present, in a first screen of a graphical user interface (GUI), a design space for a protein for an application, wherein the design space comprises a plurality of protein sequences, wherein:
each protein sequence is associated with a respective plurality of activities pertaining to the application, and
the plurality of activities comprises one or more biomedical activities, biochemical activities, or some combination thereof, and
the plurality of protein sequences is generated by a machine learning model, wherein the machine learning model uses causal inference to execute at least one of a plurality of alternative scenarios to filter a superset of protein sequences and to generate the plurality of protein sequences in the design space;
receive, via a graphical element in the first screen, a selection of one or more query parameters of the design space; and
present, in a second screen of the GUI, a solution space that includes a subset of the plurality of protein sequences, wherein each protein sequence contains the respective plurality of activities, wherein the subset of the plurality of protein sequences is selected based on the one or more query parameters;
receive, using a graphical element of the second screen, a selection of a protein sequence from the subset of the plurality of protein sequences, wherein the selection is based on an indication that the protein sequence has not been previously generated by the machine learning model; and
responsive to the selection of the protein sequence, present, in the second screen, additional information pertaining to the protein sequence, wherein the additional information comprises:
a candidate drug compound, an interaction, an activity, a drug, a gene, a pathway, or some combination thereof.
17. The computer-readable medium of claim 16 , wherein the second screen comprises:
a first portion presenting one or more color-coded clusters representing the subset of the plurality of protein sequences, and
a second portion presenting data pertaining to the subset of the plurality of protein sequences represented by the one or more color-coded clusters, wherein the data describes one or more objects associated with the subset of the plurality of protein sequences, and the one or more objects comprise a candidate drug compound, an activity, an interaction, a drug, a gene, a pathway, a physical descriptor, a characteristic, an interaction, a folding property, a wave property, a stability of modification, or some combination thereof.
18. The computer-readable medium of claim 17 , wherein the one or more color-coded clusters represent, using an energy correlation, each protein sequence in the subset, and the energy correlation comprises a correlation between each position of each protein sequence in the subset and other positions of other protein sequences in the subset.
19. A system comprising:
a memory device storing instructions; and
a processing device communicatively coupled to the memory device, the processing device executes the instructions:
present, in a first screen of a graphical user interface (GUI), a design space for a protein for an application, wherein the design space comprises a plurality of protein sequences, wherein:
each protein sequence is associated with a respective plurality of activities pertaining to the application,
the plurality of protein sequences is generated by a machine learning model, wherein the machine learning model uses causal inference to execute at least one of a plurality of alternative scenarios to filter a superset of protein sequences and to generate the plurality of protein sequences in the design space;
receive, via a graphical element in the first screen, a selection of one or more query parameters of the design space; and
present, in a second screen of the GUI, a solution space that includes a subset of the plurality of protein sequences each containing the respective plurality of activities, wherein the subset of the plurality of protein sequences is selected based on the one or more query parameters;
receive, using a graphical element of the second screen, a selection of a protein sequence from the subset of the plurality of protein sequences, wherein the selection is based on an indication that the protein sequence has not been previously generated by the machine learning model; and
responsive to the selection of the protein sequence, present, in the second screen, additional information pertaining to the protein sequence, wherein the additional information comprises:
a candidate drug compound, an interaction, an activity, a drug, a gene, a pathway, or some combination thereof.
20. An apparatus comprising:
a memory device storing instructions for presenting a graphical user interface (GUI) of a therapeutic tool; and
a processing device communicatively coupled to the memory device, the processing device is configured to:
present, in a first screen of the GUI, a design space for a protein for an application, wherein the design space comprises a plurality of protein sequences, wherein:
each protein sequence contains a respective plurality of activities pertaining to the application,
the plurality of activities comprises one or more biomedical activities, biochemical activities, or some combination thereof, and
the plurality of protein sequences is generated by a machine learning model, wherein the machine learning model uses causal inference to execute at least one of a plurality of alternative scenarios to filter a superset of protein sequences and to generate the plurality of protein sequences in the design space;
receive, via a graphical element in the first screen, a selection of one or more query parameters of the design space; and
present, in a second screen of the GUI, a solution space comprising a subset of the plurality of protein sequences, wherein each protein sequence contains the respective plurality of activities, wherein the subset of the plurality of protein sequences is selected based on the one or more query parameters;
receive, using a graphical element of the second screen, a selection of a protein sequence from the subset of the plurality of protein sequences, wherein the selection is based on an indication that the protein sequence has not been previously generated by the machine learning model; and
responsive to the selection of the protein sequence, present, in the second screen, additional information pertaining to the protein sequence, wherein the additional information comprises:
a candidate drug compound, an interaction, an activity, a drug, a gene, a pathway, or some combination thereof.
21. The apparatus of claim 20 , wherein the second screen comprises:
a first portion presenting one or more color-coded clusters representing the subset of the plurality of protein sequences, and
a second portion presenting data pertaining to the subset of the plurality of protein sequences represented by the one or more color-coded clusters, wherein the data describes one or more objects associated with the subset of the plurality of protein sequences, and the one or more objects comprise a candidate drug compound, an activity, an interaction, a drug, a gene, a pathway, a physical descriptor, a characteristic, an interaction, a folding property, a wave property, a stability of modification, or some combination thereof.
22. The apparatus of claim 21 , wherein the one or more color-coded clusters represent, using an energy correlation, each sequence in the subset, and the energy correlation comprises at least one correlation between each position of each protein sequence in the subset and other positions of other protein sequences in the subset.
23. The apparatus of claim 20 , wherein the solution space is presented as a topographical map in the GUI, wherein the topographical map comprises a plurality of indications that each represent a level of activity for a protein sequence associated with a given point on the topographical map.
24. The apparatus of claim 20 , wherein the design space is generated based on a knowledge graph pertaining to peptides and the design space is presented as a two-dimensional (2D) elevation map, a three-dimensional (3D) shape or an n-dimensional (nD) mathematical representation.
25. The apparatus of claim 20 , wherein the solution space is generated within the design space by one or more machine learning models trained to measure, based on the query parameter, a respective level of one or more of the respective plurality of activities of each of the plurality of protein sequences in the subset, wherein the query parameter comprises a sequence parameter.
26. The apparatus of claim 20 , wherein the processing device is configured to:
receive, using a graphical element of the second screen, a selection of a protein sequence from the subset of the plurality of protein sequences; and
present, in a third screen of the GUI, a candidate dashboard comprising information pertaining to the protein sequence, wherein the information pertains to a structure of the protein sequence, a correlation heatmap, experimental data, a list of probabilistic scores generated by inference models, external data related to the sequence, or some combination thereof.
27. The apparatus of claim 20 , wherein the processing device is configured to:
receive a selection of a trial configured to be performed by a machine learning model, wherein the machine learning model uses the solution space; and
receive, from an artificial intelligence engine, one or more results of performing the trial, wherein the one or more results:
provide a location of a point reached in the solution space after performing a traversal of the solution space defined by the trial, and
provide a metric of a machine learning model used by the artificial intelligence engine to perform the trial, wherein the metric pertains to one or more of memory usage, graphic processing unit temperature, power usage, processor usage, or central processing unit temperature.
28. The apparatus of claim 20 , wherein the processing device is configured to:
receive, from a graphical element of a business intelligence screen of the GUI, a target product profile, wherein the target product profile comprises pharmacology data, pharmacokinetic data, pharmacodynamic data, activity data, manufacturing data, compliance data, clinical trial data, or some combination thereof;
receive, from an artificial intelligence engine, a second subset of the plurality of protein sequences, wherein the second subset of the plurality of protein sequences is selected based on the target product profile; and
present, in the GUI, the second subset of the plurality of protein sequences.