Grant: Discovery Grants Program - Individual

Biomarkers are indicators used to objectively measure and evaluate normal biological processes, pathogenic processes or pharmacological responses to therapeutic interventions. In medical practice, biomarkers facilitate diagnosis and prognosis of disease, as well as monitoring clinical response to therapy. In environmental studies, biomarkers can be used to assess the health states of animals and plants, with respect to, for instance, the amounts of toxic materials they are exposed. Advancements in the fields of genomics and proteomics have revolutionized novel biomarker discovery with new tools for high throughput strategies. In this proposal, we present a comprehensive framework for using large amounts of genomics and proteomics data for identifying biomarkers and signatures. The proposal has three key focuses: (A) data cleansing and pre-processing; (B) biomarker and signature identification; and (C) validation with real data sets. For data cleansing and pre-processing, we consider a range of topics: (A1) Quality control for microarray data; (A2) Proteomics platform calibration; and (A3) Formation of protein groups for proteomics data. We emphasize that before data can be mined, they need to be properly cleansed and transformed. For biomarker and signature identification, we explore a number of key topics: (B1) Gene set enrichment analysis; (B2) Ensemble methods for combinatorial analysis; and (B3) Embedded clustering methods during biomarker discovery. Last but not least, for the objective of validation with real data sets, we discuss how the aforementioned tasks will be applied to two large-scale genomics and proteomics data sets funded by Genome Canada and NSERC.