This project two serves as the bookends of a hypothesis-test-verify cycle to define the shared and distinct components of neurodevelopmental diseases. For the first bookend: In the context of a large healthcare system, we use the processed electronic health records (through Natural Language ProcessingNLP) to determine the temporal course of multiple health/clinical attributes observed and/or obtained during the course of care delivery. These then are used, in collaboration with Project 4 to define temporal co-morbidity trajectories for each of these diseases and to determine the extent to which there is shared phenomenology particularly in time. For the second bookend: those genomic variants/sequences that are predicted/hypothesized to define subgroups previously characterized by clinical trajectories, or predict conventional biomarkers/clinical findings and/or (most preferably but most challenging) predict differential therapeutic responsiveness.
The project aims to model succession of environmental factors and disease phenotypes in the life of patients, using data from large clinical databases. This project focuses on complex neuropsychiatric disorders, such as autism, schizophrenia and depression.
Showing the most recent 10 out of 83 publications
