Finding Genomic Profiles of COVID-19 Phenotypes from the EHR

Abstract: As the world searches for effective treatments and potential cures for the COVID-19 pandemic, theability to consolidate data, insights, and expertise from many disparate sources will be key to fullyunderstanding the patient outcomes of the infection. Key to facilitating this type of research is a cohesive andsecure research environment that enables clinicians, researchers, data scientists, and technologists frommultiple organizations to work together with a common goal of better understanding COVID-19 symptoms,associated risk factors, and successful therapies. Building upon faculty, staff and infrastructure already in placethrough the eMERGE IV Clinical Center at Mass General Brigham, we are proposing the creation of a COVID-19 Biobank Portal to provide a foundation for building a truly collaborative environment that is compliant withpatient privacy and offers a common set of bioinformatic tools and a standardized IT approach for the stagingof data and analyses. We will do this by accomplishing the following three Specific Aims which supplement theparent grant's Aim I which is: "Polygenic risks scores will allow us to stratify eMERGE participants based ongenetic risk for common complex traits" which will focus in this supplement on risk factors for severity ofCOVID-19 illness in our biobank participants. We propose to build on our expertise to accomplish the specificaims: Aim 1: We will create a COVID-19 Centric Biobank Portal that allows general institutional use withproper research agreements in place where patient cohorts can be studied using easily assessable andtransformed data and through which genomic samples can be obtained. Aim 2: Supplement the COVID-19Biobank Portal with test results, phenotype risk factors, symptoms, and outcomes for COVID-19 which arederived from data in the electronic health record (EHR) by using natural language processing andcomputational phenotypes and by performing chart reviews to validate severity indices and clinical outcomesfound in COVID-19 infected patients. Aim 3: Genetic data (array and sequence data) will be contributed to theeMERGE IV network and to the International l COVID-19 Host Genetics Initiative (https://covid19hg.org) suchthat data can be used to calculate polygenic risk scores (PRS) for genome-wide association studies of riskphenotypes and patient outcomes and polygenic risk scores (PRS) for COVID-19 outcomes in our dataset andin collaborations with others.