This page includes detailed text about Precision Health, data access/storage and technology resources that may be helpful in preparation of grant applications.
About Precision Health
Precision Health at the University of Michigan is a campus-wide, multidisciplinary, presidential initiative funded by the Provost, the Medical School, the College of Engineering, and the School of Public Health and involves all 19 schools and colleges at the university, as well as partnering centers and initiatives. Precision Health aims to discover and validate genetic, environmental, social, behavioral, and clinical markers that influence disease prevention and health outcomes with the ultimate goals of translating these discoveries into personalized treatments and improving the health of communities around the world.
Led by faculty co-directors Sebastian Zöllner, PhD (SPH), Brahmajee Nallamothu, MD, MPH (Michigan Medicine), and Jenna Wiens, PhD (CoE), PH is housed at the North Campus Research Complex at the University of Michigan, near valuable partners, like the Institute for Healthcare Policy & Innovation (IHPI), the Data Office for Clinical & Translational Research (DOCTR), the Research Data Warehouse (RDW), and the Central Biorepository (CBR). PH researchers and staff are deeply embedded in these related offices to allow for seamless collaboration and provide staff and administrative support to research projects across the university.
Focused on providing the infrastructure and platform needed to enable science across disciplines, PH comprises five tightly linked infrastructure workgroups. These workgroups provide resources in 1) data analytics and information technology, 2) patient cohort development and datasets, 3) health implementation, 4) education and training programs for professionals in precision health science, and 5) funding and research opportunities. Through our workgroups, investigators have access to research data accessible on HIPAA-compliant servers, genomic data for MGI participants, data sciences platforms, and national networks. Managed by experts in areas such as data science, biostatistics, machine learning, biomedical engineering, genomics, anesthesiology and pain management, psychiatry, bioinformatics, and pharmacogenomics, PH provides a deeply diverse, intellectual environment that fosters collaboration.
About the Michigan Genomics Initiative
The Michigan Genomics Initiative (MGI) is a collaborative research effort among physicians, researchers, and patients at the University of Michigan with the goal of combining patient Electronic Health Record (EHR) data with corresponding genetic data to gain novel biomedical insights. Biospecimens that are collected from participants are sent to the Central Biorepository for processing, and DNA is isolated from the biospecimens. A portion of that DNA is set aside for array genotyping by the Advanced Genomics Core.
MGI is a single-health-system biobank from patients at the University of Michigan. MGI participants consent to linking genotype data to existing and future clinical data for broad research purposes, and they provide socio-demographic and lifestyle information. To date, >85,000 people are enrolled in the biobank, with ongoing recruitment of ~1,000 per month. The first ~60,000 individuals have been genotyped in four waves using customized versions of the Illumina Infinium CoreExome-24 bead array platform and imputed using the TOPMed reference. Genotyping with the Illumina GSA chip is ongoing. Most of the sample self-identifies as European American (49,605 in waves 1-4, 87%), with African American (3,223, 5.6%) and Asian American (1,324, 2.3%) next most common. MGI is part of the global biobank meta-analysis initiative (GBMI), a 21-biobank consortium with >2.6 million genotyped samples.
MGI Documents for Researchers
- Precision Health at the University of Michigan Policy on Access to Genomic, Health, and Summary Data, Biospecimens, and Re-contact of Research Participants Enrolled in the Michigan Genomics Initiative [pdf]
- University of Michigan Precision Health Policy Establishing a Michigan Genomics Partnership [pdf]
- Michigan Genomics Initiative Acknowledgment and Authorship Policy [pdf]
Protected Health Information (PHI) and Honest Broker Services
Precision Health adheres to the HIPAA Privacy Rule (Section 165.514(a) and uses the “Safe Harbor” method to code patient data with the additional nuance that all dates are shifted (where the shift preserves temporal relationship for each patient/participant). Honest Broker services under the Data Office for Clinical & Translational Research (DOCTR) ensure a highly secure conduit between protected patient information and the research end user. These services are designed to operationalize IRB and compliance policies, including assistance with data sharing both internally and with external institutions; identifying secure storage options; linking disparate datasets with a coded-identifier methodology; keeping the key for encrypted data identifiers; and data auditing services.
Data Access and Security
The University of Michigan Information Assurance (IA) Office leads IT security, privacy, identity and access, and IT policy and compliance efforts that enable the University to excel in its teaching, research, and patient care missions. Information Technology Services (ITS) works with Michigan Medicine Corporate Compliance to review its overarching HIPAA compliance program that includes documented and implemented administrative, physical, and technical safeguards. Examples include Unique User Identification, mandatory two-factor authentication, routine collection and review of authentication logs; and automatic logoff or lockout of workstations after 15 minutes of no activity.
Data and Computing Resources
DataDirect
Researchers have access to an investigator-friendly, self-service tool called DataDirect. With more than 700 unique users annually, DataDirect provides researchers real-time access to cohort counts (no IRB approval required) from EHR data, available biospecimens, and genomic data. With IRB approval, research teams can develop custom data extracts to download and analyze. For clinical trial enrollment, DataDirect has functionality to link eligibility criteria with upcoming clinic schedules to optimize recruitment.
DeIdentified Research Data Warehouse (RDW)
The Precision Health Deidentified RDW (Research Data Warehouse) offers researchers direct access to database tables that house the clinical structured data of 4M+ Michigan Medicine patients. The Deidentified RDW can be queried in the Precision Health secure enclave through SQL, Python, and RStudio.
Armis2
The Armis2 high-performance computing (HPC) environment is a Linux-based high performance computing cluster used for analysis of sensitive data administered by Academic Research Computing . Precision Health also has a private set of six nodes on Armis2. Each node has eight (48 total) RTX2080Ti GPUs and large volumes of fast local storage and can see all data and software provided on Armis2. These nodes are optimized for machine learning/AI, computer vision, molecular dynamics, and any other GPU-accelerated workload. Precision Health also has a dedicated partition with the campus-wide Armis2 computing nodes, which pulls from the general pool of GPU and computing resources.
Secure Enclave Services (SES)
SES is a private cloud environment that provides high-performance, secure, and flexible computing environments enabling the analysis of sensitive datasets restricted by federal privacy laws, proprietary access agreements, or confidentiality requirements. SES is administered by the University of Michigan Academic Research Computing group.
Turbo
Turbo is a high speed network storage platform offered by ARC-TS for research purposes. It can be mounted in other high performance computing environments, including YBRC and Armis2. Turbo is administered by the University of Michigan Academic Research Computing group.
Encore
Developed at the Center for Statistical Genetics, Encore is a self-serve, web-based analysis tool for running genome-wide association studies (GWAS). Investigators are able to run GWAS in a point-and-click interface without needing to directly manipulate or “touch” the genetic data. Only a phenotype file is needed to build a GWAS model with SAIGE (genetics analysis software), launch and monitor job progress, and interactively explore results.
