DEPARTMENT OF APPLIED MATHEMATICS 100 Email kin-yau.wong@polyu.edu.hk Qualification BSc (The University of Hong Kong) MPhil (The University of Hong Kong) PhD (The University of North Carolina at Chapel Hill) ORCID ID 0000-0001-9066-1619 Dr WONG Kin-yau Alex Associate Professor Research Overview My research focuses on the development of effective, computationally efficient, and theoretically sound statistical methods for the analysis of public health and medical studies. My areas of interest can be broadly classified into survival analysis and integrative analysis of omics data. In public health and medical studies, investigators are often interested in time-to-event outcomes, such as the times to disease events or death of patients since diagnosis, and their relationship with treatment and other personal risk factors. Due to practical limitations, a subject’s exact event times are potentially unobserved but are only known to be after some time point or within a given time interval. Other common complications in real studies include the presence of a cure, that is, a phenomenon where subjects may not be susceptible to the event of interest, and dependent data, where multiple event times may be associated. My work in this area is centred on building flexible semiparametric models and developing valid estimation methods for data subject to different types of complications encountered in practice. Another area of interest is the integrative analysis of omics data. Recent technological advances have made it possible to collect multiple types of omics data, such as DNA alteration, RNA expression, and protein expression, on a large number of subjects. The availability of such detailed molecular data is valuable for the study of disease mechanisms and subsequently the development of effective diagnostic methods and treatments. In our work, we focus on the development of models for multiple potentially highdimensional omics features and disease outcomes that capture the intricate association among the variables (see the figure below). We aim at efficiently leveraging information from various types of omics features to identify biologically relevant features and to build prediction models for disease outcomes. Representative Publications • Ann. Stat. 2022, 50, 487-510 • Biometrics 2022, 78, 165-178 • Stat. Med. 2021, 40, 2400-2412 • Genome Biol. 2019, 20, 52 Association structure of different types of omics features
RkJQdWJsaXNoZXIy Mjc5OTU=