skip to content

Cambridge Cardiovascular


Assistant Professor of Health Data Science


Sam Lambert is an Assistant Professor of Health Data Science within the Cardiovascular Epidemiology Unit (CEU) of the Department of Public Health & Primary Care. Before his postdoctoral work in Cambridge, Sam completed his B.Sc. in Biological Sciences at the University of Guelph and a Ph.D. in Molecular Genetics at the University of Toronto. He joined the CEU as a research associate in 2019 under the supervision of Dr. Michael Inouye, funded by a Canadian Institutes of Health Research fellowship until starting his current position in July 2023. Sam is also a Fellow of Churchill College and a visiting researcher at the European Bioinformatics Institute (EBI).


The primary theme of research in my group is to understand and predict patterns of multimorbidity (two or more chronic conditions affecting an individual) that involve common cardiometabolic diseases. We investigate these questions in deeply phenotyped cohorts of individuals with multi-omic measurements (genetic, proteomic, metabolomic, and other biomarkers) and linked health record data. With these data, we can identify the molecules and pathways that drive differential susceptibility to multimorbidity patterns by applying genetic techniques (e.g., genome-wide association studies, polygenic scores, mendelian randomization) and other data science methods. We are also interested in whether predictions of multimorbidity risk can be made to prevent the onset of comorbidities in people with an initial condition by combining phenotypes extracted from health records (longitudinal measurements, medication usage), genetic data/PGS and methods capable of considering the risk of multiple diseases at once.

The second focus of the group is related to open science and FAIR (findable, accessible, interoperable, and reusable) data. In collaboration with Dr. Michael Inouye (Cambridge-Baker Systems Genomics Initiative), I lead the development of the Polygenic Score (PGS) Catalog ( which aims to increase the transparency and reproducibility of PGS by distributing the information necessary for both research and translational uses of PGS for a wide range of diseases and traits. To make PGS data more useful we also perform benchmarking of PGS performance in diverse cohorts to provide consistent descriptions of their predictive ability and develop tools to make PGS calculations more reproducible ( This work is in collaboration with partners from the GWAS Catalog at the European Bioinformatics Institute and supported by Health Data Research UK (HDR-UK) and the USA National Institutes of Health (NIH).


Other publications: 

See Google Scholar for a list of publications.