Deep Mutational Scanning of Monogenic Diabetes Genes to Facilitate Precision Diagnostics for Diabetes

PROJECT SUMMARY/ABSTRACT Diabetes is a global health challenge and affects around 463 million people worldwide. For up to 5% of these individuals their diabetes is due to a defect in a single gene. These monogenic varieties of diabetes provide important opportunities for precision medicine altering the first line treatment for patients, providing information

on prognosis and risk for family members. Our inability to interpret DNA sequence variation results in uncertainty regarding the impact of a DNA change on protein function leading to variants of unknown significance (VUS) which are a barrier to precision medicine. A proposed solution is the deployment of multi-variant assays of

effects (MAVES) which generate a comprehensive catalogue of the effect of DNA sequence on protein function which can then aid clinical variant interpretation and provide important information on protein structure and function. The choice of assay and its alignment with gold-standard low throughput approaches is paramount for

accurate variant interpretation. The overarching aim of our research program is to generate comprehensive maps of variant effects for genes involved in monogenic diabetes to deliver precision diagnostics to enable precision medicine. We will capitalize on our expertise in gold-standard assays for variant characterization in monogenic diabetes

genes and alignment with the Clin Gen Monogenic Diabetes Expert Panel. We will deploy state-of-the-art microfluidic platforms and use authentic human cell models coupled to gene and disease relevant assays to determine the effects of all missense variants which can be generated by a single nucleotide substitution for the

two most common causes of monogenic diabetes, defects in the key glycolytic enzyme glucokinase (GCK) and the transcription factor hepatocyte nuclear factor 1 alpha (HNF1A). We will achieve this by [1] Quantitatively characterizing the `functional enzymatic signatures' across 1000s of GCK variants using High-

Throughput Microfluidic Enzyme Kinetics (HI-MEK); [2] Determining HNF1A transcription factor function at scale using Simultaneous Transcription factor Affinity Measurements via Microfluidic Protein Arrays (STAMMP) to determine effects on DNA binding and coupling this with a high-throughput pooled screen for insulin secretion

in human beta cells and; [3] working with international leaders in monogenic diabetes to Integrate our variant maps into clinical diagnostics through variant curation & the developing improved prediction models. Outcome: We will deliver a roadmap for design, validation, and implementation of MAVES to assess the impact

of protein coding variants in any medically actionable gene relevant to diabetes risk.