Longitudinal Histories as Predictors of Future Suicide Attempts
Suicide is a leading cause of death, especially among young people. Early and accurate identification of individuals at elevated risk for suicide attempts and death is critical for developing and implementing effective strategies to prevent suicidal behavior. This study aims is leveraging longitudinal data in electronic health records to develop and validate predictive algorithms that might inform suicide risk assessment and to evaluate the potential of a clinical decision support tool based on this work. This research has been supported by the Tommy Fuss Fund.
Improved multifactorial prediction of suicidal behavior through integration of multiple datasets.
Through funding provided by the NIMH, we are expanding our efforts to leverage health data for suicide risk prediction in large-scale healthcare systems. This effort includes applying a a range of machine learning methods, integrating multiple data sources, applying natural language processing approaches, and developing novel methods to incorporate time-dependent risk profiles.
Development and validation of an electronic health record prediction tool for first-episode psychosis
The development of a scalable, generalizable EHR-based screening tool for first episode of psychosis (FEP) may enable earlier interventions of psychotic disorders and decrease duration of untreated psychosis (DUP), a significant factor impacting the prognosis of psychotic disorders. In collaboration with colleagues at Boston Children’s Hospital (BCH) and Boston Medical Center (BMC), our team is developing and validating tools for prediction and screening of FEP using healthcare system data from diverse clinical populations. We are applying a range of machine learning methods to optimize model performance. In addition, we will engage stakeholders in prototyping a clinician-facing EHR-based risk screening tool for FEP.
PsycheMERGE: Leveraging electronic health records and genomics for mental health research
The PsycheMERGE consortium grew out of the eMERGE network and includes investigators from multiple institutions with EHR-linked biobanks. PsycheMERGE aims to leveraging health system data, biomedical informatics and genomics to facilitate precision medicine research in psychiatry. This work includes the development and validation of EHR-based phenotyping algorithms, cross-disorder genomic studies, and applications of EHR and genomics for prediction of disorder risk and other clinical outcomes.
All of Us Research Program
The All of Us (AoU) program is a component of the Precision Medicine Initiative (PMI), which was launched in 2016 to facilitate precision medicine research across all areas of health and disease. the individual factors that influence health such as their environment, lifestyle, and unique biology. The All of Us program will enroll and partner with more than one million participants reflecting the broad diversity of the U.S. to better understand the role that individual differences in biology, environment, and lifestyle play in the prevention, diagnosis and treatment of disease. The New England Precision Medicine Consortium is a component of the All of Us Research Program and a collaboration of Partners Healthcare and Boston Medical Center. https://www.joinallofus.org/en/newengland
eMERGE Phase III Clinical Center at Partners HealthCare
The primary goal of the eMERGE Network is to develop, disseminate, and apply approaches to research that combine biorepositories with electronic medical record (EMR) systems for genomic discovery and genomic medicine implementation research. In addition, the consortium includes a focus on social and ethical issues such as privacy, confidentiality, and interactions with the broader community. The eMERGE III Clinical Center at Partners HealthCare is leveraging the Partners Biobank to define the phenotypic impact of common and rare genetic variation on a range of diseases and to develop, implement, and evaluate procedures for return of genetic results on selected variants to participants.
Determining the Effects of Missense Variant in SLC39A8 on Serum Metals and Protein Glycosylation in Schizophrenia
Led by Dr. Robert Mealer, we are examining the biochemical and functional genomics of variants in SLC398A8, which encodes for a zinc/manganese transporter protein and which has been associated with schizophrenia and a range of other medical conditions.
Partners Healthcare Training Program in Precision and Genomic Medicine (T32)
The Partners Training Program in Precision & Genomic Medicine is designed to provide rigorous, interdisciplinary training of postdoctoral scientists in the translational of genomic, clinical and computational sciences that are driving a new era of precision and genomic medicine. https://cgm.massgeneral.org/training-program/