Description:  Leadership Position Opening: Director of Center for Precision Medicine in the Department of Biomedical Informatics and Vice President of Personalized Medicine at Vanderbilt University Medical Center

Position Details: Vanderbilt University Medical Center (VUMC) is seeking a senior faculty member, professor or associate professor, with national recognition as an expert in precision medicine to lead our Center for Precision Medicine. This position is focused on leading the development and implementation of methods and technologies to translate genetic/genomic, bioinformatic, clinical, and patient-centered discoveries from research to clinical environments. VUMC is particularly interested in candidates with experience managing a grant portfolio focused on hypothesis generation research with expertise in translational informatics and genomic medicine methods and a proven track record of collaboration across diverse institutional centers and departments.  

The successful applicant will be able to immediately engage in ongoing grant-funded work to develop state-of-the-art precision medicine. As a part of a highly productive, highly supportive interdisciplinary team of researchers, data scientists, informatics leaders, and clinical and genomic experts, this person will utilize biomedical research and computational methods to develop and integrate tools into real-world enterprise-wide healthcare settings. The faculty appointment will reside in the Department of Biomedical Informatics, with the expectation of cross-organization collaboration and leadership in the applied field of precision medicine.  Candidates for this position should hold either a doctorate in informatics, computer science, genomics, or a related discipline or a medical degree with demonstrated experience/training in informatics or genomics.

Applicants for this position should submit a cover letter, CV, and the names and contact information for a minimum of three references. The cover letter should describe the applicant’s relevant qualifications and experience, as well as brief statements of research and teaching interests. All application materials should be sent to paul.a.harris@vumc.org and brought to the attention of Paul A. Harris, Ph.D. Professor of Biomedical Informatics, 2525 West End Avenue – Suite 1060, Nashville, TN 37203 Office: 615-322-6688.

Information about the Department: 

DBMI is an internationally recognized leader in biomedical informatics, with strong relationships with academic departments in the university and the clinical operations of a large medical center. DBMI has an emphasis on interdisciplinary research, with a focus on health information technology development, clinical knowledge and workflow modeling, implementation sciences, machine learning and clinical decision support, biomedical data analytics (clinical, genomic, and proteomic), and natural language processing, privacy, and security.  

Among more than 50 faculty are six members of the National Academy of Sciences and nearly 20 members of the American College of Medical Informatics. DBMI faculty holds additional appointments in various departments, including anesthesiology, biochemistry, biomedical engineering, computer science, hematology and oncology, human genetics, internal medicine, and pediatrics. 

Center For Precision Medicine and Resources

The Center for Precision Medicine (CPM) at Vanderbilt University Medical Center is a multidisciplinary team consisting of basic research investigators, clinicians, data scientists, application developers, and support personnel – 23 faculty, and 28 staff and trainees – whose objective is to build a research platform that amasses, extracts and curates meaningful information from patient healthcare records and integrates with genetic information derived from a DNA repository patterned after BioVU. The CPM develops and applies advanced bioinformatics tools to identify genotype/phenotype associations and genotype/treatment relationships in order to facilitate discovery and signally improve healthcare outcomes through the personalization of diagnostics, treatment, and disease prevention. 

Precision Phenomics, including PheWAS: The CPM team has significant experience mining EMR records, including the development of novel tools and methods for defining cohorts and extracting clinical data. The PheWAS methodology, for example, and its use as both a replication tool for GWAS and a tool for discovery of genetic pleiotropy were first pioneered by Dr. Josh Denny and the VUMC CPM team, using EMR data linked to BioVU. Detailed clinical facets of the disease are often captured in unstructured narrative documents, such as discharge summaries, clinic notes, radiology reports (e.g., bone erosions), or clinical reports (e.g., echocardiograms or catheterization reports). A suite of Natural Language Processing (NLP) tools have been developed to extract phenotypic attributes in order to identify patients and cohorts of interest. These tools include a general-purpose NLP engine (KnowledgeMap concept identifier, KMCI), a machine-learning-based clinical note section tagger (SecTag), and a medication extraction algorithm (MedEx). The systems have been used to process >60 million documents at Vanderbilt and other institutions. The ever-growing list of completed and validated phenotypes here through the eMERGE Network (VUMC serves as an eMERGE site and as the eMERGE Coordinating Center) and other projects shows the degree to which we are experienced with mining the EMR for the rapid development of accurate, reliable, and precise case and control counts for estimating patient accrual. To coincide with our phenotype development and validation efforts, CPM has also developed and deployed the Phenotype Knowledgebase (PheKB, https://phekb.org/) as a collaborative environment for building and validating electronic algorithms to identify patient cohorts or characteristics within health data. PheKB has flexible integrated data aggregation and validation tools to aid in harmonizing data across diverse sites. PheKB was functionally designed to enable efficient workflow and has purposefully integrated tools and standards that guide the user in navigating from early-stage development to public sharing and reuse. PheKB has tools to enable multisite collaboration for algorithm development, validation, and sharing for reuse. PheKB is relevant for the storage and reuse of algorithms and for identifying populations in future PMI prospective research. In Nov 2016, the VUMC Phenotyping and PheWAS shared resources core was established. To date, the core has supported >85 research groups with customized phenotyping, PheWAS, and GWAS analyses using the tools and resources developed by the CPM research team.  

-------------------

Vanderbilt University Medical Center is an equal-opportunity, affirmative-action employer. Applications from women and members of underrepresented minority groups are strongly encouraged. For more information, please visit the DBMI web site at http://medschool.vanderbilt.edu/dbmi/. 

-------------------

Vanderbilt University Medical Center and School of Medicine

VUMC is a comprehensive health care facility dedicated to patient care, research, and the education of health care professionals. Its reputation for excellence in each of these areas has made Vanderbilt a major patient referral center. Translational research on the causes and treatment of disease is the focus of discovery at Vanderbilt. Vanderbilt faculty and medical house staff that provide clinical care and participate in research programs. Vanderbilt has more than 450 research laboratories; support for competitive research grants from all external sources totaled more than $657 million in FY 2015. Approximately 2,555 graduate-level students are pursuing advanced degrees in health sciences. VUMC is home to the region's only Level I Trauma Center, a dedicated Regional Burn Center, and one of only 45 Comprehensive Cancer Centers in the nation. VUMC hospitals, clinics, physician practices, and affiliates cover 172 counties, eight hospital systems, and 48 hospital locations. VUMC is regularly named among the nation’s top 100 hospitals by Thomson Reuters. The Vanderbilt University Hospital (VUH), The Monroe Carell Jr. Children’s Hospital and the Vanderbilt Clinic are all part of VUMC. The clinical mission reflected in these resources drives our motivation to pursue translational science and improve patient care.

Vanderbilt University School of Medicine: The School of Medicine, originally part of the University of Nashville, was incorporated into Vanderbilt University in 1874 and awarded its first Vanderbilt medical degrees in 1875. Since the inception of the School of Medicine, a Vanderbilt medical education has been held in high esteem among its peer institutions, and that legacy continues today. Biomedical research at the School of Medicine has long been recognized for its contributions to the advancement of medicine. The School of Medicine claims two Nobel Laureates, Earl Sutherland Jr., in 1971, for his discovery of the metabolic regulating compound "cyclic AMP," and Stanley Cohen, in 1986, for his and a colleague’s discovery of epidermal growth factor. Eight of the School of Medicine’s basic science departments – physiology, pharmacology, pediatrics, biochemistry, radiology, anesthesiology, medicine and cell and developmental biology – rank in the top 10 in the country in terms of competitive NIH funding. The School of Medicine currently has 1,833 faculty and more than 600 students. From 2000-2014, Vanderbilt University School of Medicine (VUSM) had the fastest growth in NIH funding among all academic medical centers (17.8%).

Research Infrastructure

In addition to the highly collaborative and high-quality academic health center, VUMC has invested in world-class research infrastructure.  

Vanderbilt Institute for Clinical and Translational Research (VICTR): The CTSA housed at Vanderbilt (VICTR) is in its 10th year, with a 5-year renewal recently awarded.  The program’s initial goal in 2006, when first founded, was to assist institutions in creating a “novel and integrative academic home for clinical and translational science.” VICTR was created to be the single home for translational research at Vanderbilt. The foundation for VICTR was built upon an integrated, comprehensive, informatics-driven research and administrative infrastructure that served as a launchpad for numerous transformative, research-enabling programs and resources such as Studios, pilot funding vouchers, BioVU, REDCap, ResearchMatch, StarBRITE, and community engagement initiatives. Using this infrastructure as the cornerstone, we were able to leverage, grow, and refine our resources, programs, and operations during the second funding cycle while applying them to enhance the quality and efficiency of the research conducted. Moreover, we expanded the reach of our programs at a local, national, and international scale. Today, VICTR is a highly functional and integrated translational research infrastructure that has raised the quality and scientific rigor of the research conducted at Vanderbilt and longstanding partner Meharry, the nation’s oldest historically black academic health science institution. VICTR is now intimately woven into the fabric and impacts every facet of clinical and translational (C&T) research, including the conduct of pragmatic clinical trials that fuel the VUMC Learning Health System. Please see https://victr.vumc.org/ for more information about our clinical and translational science infrastructure.

Vanderbilt’s Electronic Health Record (EHR) system: In November 2017, Vanderbilt fully implemented Epic for all inpatient and outpatient medical records and clinic workflow, pharmacy management, and hospital billing. For the operational database server, where end-user response time is critical, the EHR uses Epic’s Chronicles Extended RDBMS. Chronicles use the fast data access layer of InterSystems Caché for storage and concurrency control, and all higher-level database functionality is built within Chronicles. Chronicles pre-joins data based on expected access patterns and provides a native over-time orientation that is a good match for the structure of patient data. This allows for massive scalability while still providing a relational view. To help ensure optimal performance for users, Epic uses a configuration in which retrospective reports are on a database separate from the production database. The reporting database (or the EDW) is used for analytical and statistical reporting. This database is populated through regular (typically nightly), incremental extracts of production data. This data is primarily comprised of patient encounters and financial transaction data.

My Health at Vanderbilt patient portal: Initially implemented in 2003, Vanderbilt’s the My Health at Vanderbilt (MHAV) is one of the oldest and best-adopted portals in the country. MHAV is currently implemented using the Epic MyChart platform as a secure personal health record system designed for patients to view and manage their healthcare. My Health at Vanderbilt includes common patient portal functionality: it enables secure communication between patients or their proxies and their doctors, allows patients to access personal test results, and provides a mechanism for patients to learn about their health conditions. In addition, MHAV allows patients to review and suggest changes to their lists of problems, medications, and allergies, to check-in for appointments early, and to schedule new appointments. Currently, more than 400,000 total patients have ever registered for MHAV, and it is regularly used by 42% of adult patients across the institution. In some clinics providing ongoing chare for chronic diseases such as cancer, MHAV adoption is over 80%. By design, MHAV can be readily utilized to facilitate rapid primary data collection in a large patient population. Information about MHAV is provided to demonstrate that we have developed and applied the expertise to communicate meaningfully with patients as a part of their healthcare, including the integrated processes, interface, and privacy regulations associated with doing so. The medical center emphasizes MHAV as a central technology for its goals around patient engagement, clinical efficiency, and paperless clinics.

BioVU: BioVU, the Vanderbilt DNA Databank, is a repository of de-identified DNA extracted from discarded blood collected during routine clinical testing Although BioVU initially used an opt-out consent mechanism, individuals currently choose to opt-in through kiosks at clinic check-in that explicitly discuss information sharing. BioVU has created a centralized resource for investigating genotype-phenotype associations that have enabled innovative research at large scale. Biospecimens within BioVU are matched with corresponding clinical and demographic data derived from a de-identified research database, the Synthetic Derivative (the “SD” described below). It is overseen by a comprehensive governance structure. BioVU contains nearly 250,000 unique samples as of January 2019 and has generated 256 studies resulting in 361 publications. BioVU’s time and cost efficiency have been documented. (Bowton et al Sci Trans Med 2014) BioVU has always required data redeposit, enabling reuse by others. BioVU data deposit into dbGaP also enables derivative research worldwide: data from BioVU have been used >610 times through dbGaP. 

Synthetic Derivative (SD): The SD contains all clinical information in the EHR and its associated entry-order relational database but is stripped of personal identifiers and modified in other ways to improve data reusability. The name ‘synthetic derivative’ comes from both alterations (e.g., date shifting to mask actual dates, which protects against re-identification) and extractions (e.g. of textual and structured information that is identifiable). The SD currently contains >3 million records with highly detailed longitudinal data for over one million subjects. The SD has been used for 590 studies to date, including those by new and junior faculty, as well as those without any technical proficiency. The de-identification methodology is based primarily on the systematic removal of the fields that are specified in Section 164.514 of the HIPAA privacy rule; the centralized de-identification of the entire EHR promotes privacy for patients. The SD contains data beginning in the early 1980s; data since 2005 include nearly all inpatient and outpatient billing codes, laboratory values, reports, and clinical documentation, almost all in electronic formats available for searching. The database incorporates structured and unstructured data from multiple sources including diagnostic and procedure codes (ICD-9, ICD-10, and CPT); basic demographics (age, sex, race); text from clinical care including discharge summaries, nursing notes, progress notes, history and physical; problem lists; multi-disciplinary assessments; laboratory values; ECG diagnoses, clinical text and electronically derived trace values; and medication data. The SD can be used for searching and aggregating sets of cases for genomic analysis or as a stand-alone clinical research resource. As a byproduct of the clinical system, the SD contains phenotypic information that can be mined for the valid development of cases and controls with no additional study procedures. All clinical data are updated regularly to include patients new to VUMC, and therefore the SD, and to append new data to clinical records of existing patients as they continue to access care. Thus, the resource is entirely suitable for mining information relative to disease progression over time.

The Research Derivative (RD): The RD is a database of clinical and administrative data developed to enable clinical research. The RD brings together data from multiple sources, including billing codes, patient demographics, lab results, medications, and clinical narratives from over five different health information systems. The data has been structured to maximize feature searching and phenotype identification. Natural language processing and other informatics methods have been applied to transform unstructured data into information critical to electronic phenotype identification. The resource is well suited for rapid, efficient extraction of clinical data on a defined cohort using specific tests or phenotypes as inclusion criteria to deliver identified or de-identified datasets, recurring reports, and up-to-date counts of subjects meeting inclusion criteria. Instead of labor- and cost-intensive manual chart review that is often required to identify and study a targeted patient population that normally occurs over months, the RD allows database analyst-driven extraction of a specified clinical dataset on the scale of hours to days.

Vanderbilt Coordinating Center (VCC): The VCC has been coordinating multi-center national and international clinical trials since its inception in 1985 and was recently expanded and transitioned to a full-service Vanderbilt Shared CORE Resource, with a transparent pricing structure. VCC provides service in support of the development, design, and conduct of clinical trials ranging from 2-900 participating sites. The VCC can provide comprehensive study management support for investigator-initiated or industry/foundation sponsored trials, similar to that provided by a contract research organization or VCC customers can choose from a menu of services. Services include deliverable/timeline-based project management; study design and protocol development; database design and management (data dictionary, query, query resolution); remote and on-site monitoring; financial management (budget development/negotiation, financial tracking, invoicing); document development/management; investigator and site identification; qualification; initiation; pre-randomization participant eligibility review; randomization services; regulatory communications (FDA, IND/IDE, audit preparation, central IRB); report generation; data safety monitoring board management (recruiting members, meeting management, data report generation; documentation, communication including SAE reporting/review; medical monitoring; clinical trial material management (oversight for distribution, documentation, disposal); and medical writing.

Cross-departmental collaboration opportunities at Vanderbilt:  Data Science Institute, the Institute for Medicine and Public Health, the Health Data Science Center, the Department of Biostatistics, Genetics Institute, Institute for Clinical and Translational Research,  and more.