Posts classified under: DBDS

Maya Mathur is an interdisciplinary statistician whose research develops methods for sensitivity analysis and for evidence synthesis, particularly meta-analysis. Current focuses include developing methods for the analysis of multisite replication studies, methods for assessing and correcting for publication bias, and methods for synthesizing replication studies with existing literature. Her substantive research focuses on behavior and health and the experimental cognitive sciences; for example, her most recent empirical direction focuses on behavioral interventions to reduce meat consumption.

Research in the Mallick lab centers on developing and applying multi-scale systems approaches to enable personalized, predictive medicine in cancer. Specifically, we are developing computational methods and experimental techniques to identify diagnostic and prognostic circulating biomarkers. Biomarker-based approaches to detect cancers as early as possible and to personalize treatment are envisioned to radically improve patient outcomes and reduce healthcare costs. Within our multi-scale framework, one can consider biomarkers to be host-scale variables that inform tumor and cell-scale phenomena. Our approach to marker discovery begins with the development of molecular/cellular-scale models that attempt to describe how cells are likely to behave in response to endogenous (mutation) or exogenous perturbation (therapeutics). At the tumor-scale, we are investigating tumor heterogeneity and evolution. Recently, we have been interrogating the role of tumor-microenvironment in directing tumor evolution. At the host-scale, we are attempting to model the relationship between the tumor and the circulating proteomes to help inform biomarker candidate selection. Together, these inquiries will enable us to better understand cancer and to enable rational, model-driven approaches to biomarker discovery.

Our laboratory develops multidisciplinary engineering-based systems to study, target, and rewire cellular and multicellular circuits at scale, focusing on tumor immunology. We integrate genetic engineering tools with single cell, imaging, and spatial sequencing to perform massively parallel experiments, and use/develop machine learning and statistical inference to go from data to mechanisms and identify combinatorial effects within and across cells. Leveraging our multidisciplinary approach and capabilities we aim to identify immunomodulating interventions at an accelerated pace and lay the foundation for advances in disease diagnosis, treatment, and prevention.