Investigating the functional impact of genetic variants in the human proteome

PROJECT SUMMARY The genomic difference between individuals is estimated to be approximately 0.5% of base pairs, but this small diversity can drastically affect the interpretation of both genome- and downstream proteome-level data. While most germline variants between individuals are neutral, there are thousands of known genetic disorders in humans linked to specific germline variants. Similarly, somatic variations are a central hallmark of cancer, where mutations may cause marked phenotypic consequences in tumors as so-called “driver mutations”, but most genetic variants in tumors have little or no observable effect. Our lab is interested in understanding the effects of germline and somatic variants at the proteome level and studying changes in variant abundance over time. We believe a greater understanding of which genetic variants have molecular phenotypes will lead to improved understanding of clinically actionable features of diseases. Recent advances in proteomics have yielded a data independent acquisition mass spectrometry (DIA-MS) as a powerful new tool for quantitative studies. We have demonstrated that it is possible to accurately and consistently measure genetic variant-containing peptides (GVPs) with DIA-MS. In this proposal we will develop new bioinformatics tools to interpret GVPs using DIA-MS both to improve our sensitivity and specificity. By working with clinical collaborators to study genetic variants in amyloidosis, we hope to improve both typing and outcomes in patients carrying specific genetic variants. Furthermore, focusing on Transthyretin, the most common amyloidogenic protein, we will use mutagenized libraries to study variants in high-throughput. With this data, we will class the massive number of observed Transthyretin variants of unknown significance as either likely benign or likely pathogenic. This protein is an ideal candidate for demonstration because several Transthyretin variants have known clinical outcomes, allowing us to validate our approach. Taken together, these bioinformatic and analytical tools, coupled with validation in a disease with both known and unknown biology driven by genetic variants, will help us build powerful tools to study genetic diseases from a new proteomic viewpoint.