What We Do


The work of the MezeyLab is driven by the following question:

How do we optimally use computational statistics and machine learning to produce actionable insights and high value predictions from big biological data? 

Our approach is to focus on combinations of Biological Systems, Data Types and Analysis Methods where precise inferences can be achieved. Most of our time is spent developing and applying computational statistics and machine learning methods for analyzing big biological data types - mostly genomic data but increasingly image, clinical, and electronic health data types - in collaboration with our many partners. The scientific and application areas we work in include human genome variation, the genetic basis of diseases and other phenotypes in humans and beyond, genomic analysis of disease physiology, biomarker development ranging from disease risk prediction to diagnostics for cancer detection, and treatment development, from identifying drug targets to work on gene therapies.

For more specific information, please see pages describing Our Work, Our Research, the Data Types we work with, Our Methods, and Our Publications.