Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Discovery Grants Program - Individual

University of Toronto

$42,000.00

Towards explainable AI methods in integrative computational biology

EF 2024-25

Résumé
Without proper data management, AI algorithms cannot be effectively trained, validated and successfully applied to targeted applications. While many algorithms tolerate noise - rich, annotated and connected data (typed graphs) provide better starting point for machine learning and data mining, enable explainable AI, achieve higher accuracies in prediction, and enable more relevant models. Importantly, AI can help improving data themselves by improving organization, expanding coverage and annotation with predictions.  We propose to develop a scalable, distributed architecture to store and analyze a comprehensive database of experimentally detected, orthologous, and predicted tissue- and condition-specific interactions among biological entities. We will improve and extend our association mining algorithm, FpClass, a machine learning system for weighted link prediction between interacting entities. We will extend the method to predict links of various types between genes and proteins under the integrated graphical model in sequence, tissue, and condition-specific contexts using the Turing Complete and performance optimized graph query language Gremlin. Ontologies will be used to support multiple viewpoints and contexts. Graph theory and databases will provide necessary infrastructure and complex pattern inference platform. Machine learning and data mining will contribute new models, while probabilistic modeling will support handling incomplete, contradictory and ambiguous information. We will test and validate developed algorithms and integration platform using publicly available datasets, and in turn share them freely (https://www.cs.toronto.edu/~juris/jlab/tools.html). Together with deep learning, these approaches are revolutionizing application areas since their performance frequently exceeds domain experts' abilities and biological assay sensitivity and false discovery rates. We are also in the process of expanding and integrating World Community Grid data into an integrated, global data integration portal to increase the insights and make the data more broadly usable, using the proposed algorithms and tools on the biomedical part of the data sets. This collection of 19 years of scientific data from the World Community Grid projects (https://www.worldcommunitygrid.org/research/projects.s?proj=comp), in total over 1.5PB of data on climate, new materials, cancer and other diseases, viruses and basic biology. The proposed research will advance computational approaches with the focus on biomedical data integration, prediction and modeling. An important function of this proposal is the training of bioinformatics professionals and data scientists for whom remains a great demand. Importantly, with our World Community Grid platform (https://www.worldcommunitygrid.org), we have available sufficient computing resources, and the ability to expand training via high school outreach program to advance STEM and EDI goals.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data

Tel que publié

Application id
RGPIN-2024-04314
Name nom
Jurisica, Igor
Department département
Schroeder Arthritis Institute
Organization id
31
Institution établissement
University of Toronto
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
42000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1507
Committee name en
Computer Science
Committee name fr
Sciences informatiques
Area of application code
1201
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Life sciences (including biotechnology)
Area of application fr
Sciences de la vie (y compris biotechnologie)
Research subject code
2709
Research subject group en
Information technology
Research subject group fr
Technologie de l'information
Research subject en
Other computing methods
Research subject fr
Autres méthodes informatiques
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Towards explainable AI methods in integrative computational biology
Keywords
bioinformatics; integrative computational biology; data mining; machine learning; graph theory; knowledge management
Application summary
Without proper data management, AI algorithms cannot be effectively trained, validated and successfully applied to targeted applications. While many algorithms tolerate noise - rich, annotated and connected data (typed graphs) provide better starting point for machine learning and data mining, enable explainable AI, achieve higher accuracies in prediction, and enable more relevant models. Importantly, AI can help improving data themselves by improving organization, expanding coverage and annotation with predictions.  We propose to develop a scalable, distributed architecture to store and analyze a comprehensive database of experimentally detected, orthologous, and predicted tissue- and condition-specific interactions among biological entities. We will improve and extend our association mining algorithm, FpClass, a machine learning system for weighted link prediction between interacting entities. We will extend the method to predict links of various types between genes and proteins under the integrated graphical model in sequence, tissue, and condition-specific contexts using the Turing Complete and performance optimized graph query language Gremlin. Ontologies will be used to support multiple viewpoints and contexts. Graph theory and databases will provide necessary infrastructure and complex pattern inference platform. Machine learning and data mining will contribute new models, while probabilistic modeling will support handling incomplete, contradictory and ambiguous information. We will test and validate developed algorithms and integration platform using publicly available datasets, and in turn share them freely (https://www.cs.toronto.edu/~juris/jlab/tools.html). Together with deep learning, these approaches are revolutionizing application areas since their performance frequently exceeds domain experts' abilities and biological assay sensitivity and false discovery rates. We are also in the process of expanding and integrating World Community Grid data into an integrated, global data integration portal to increase the insights and make the data more broadly usable, using the proposed algorithms and tools on the biomedical part of the data sets. This collection of 19 years of scientific data from the World Community Grid projects (https://www.worldcommunitygrid.org/research/projects.s?proj=comp), in total over 1.5PB of data on climate, new materials, cancer and other diseases, viruses and basic biology. The proposed research will advance computational approaches with the focus on biomedical data integration, prediction and modeling. An important function of this proposal is the training of bioinformatics professionals and data scientists for whom remains a great demand. Importantly, with our World Community Grid platform (https://www.worldcommunitygrid.org), we have available sufficient computing resources, and the ability to expand training via high school outreach program to advance STEM and EDI goals.