This new T32 proposal will support graduate students and postdoctoral fellows pursuing integrated data science training in cardiovascular (CV) medicine at UCLA. Integrated data science training programs are very limited in today?s CV biomedical community. Data science challenges facing CV medicine are in many ways unique and have a long historic track record of information and data. CV diseases are chronic, heterogeneous disorders that exhibit distinct temporal profiles combined with multi-organ alterations, necessitating novel analysis platforms that integrate findings across expanded continuums of diverse information (e.g., text, imaging, omics). The complexity and size of CV datasets have pushed computational approaches to their limits, thus attenuating the rate for adding value in CV medicine. To this end, there is broad consensus that we must merge data science with CV medicine. The creation of a next-generation workforce having more advanced understanding of data science tactics for addressing real-world CV problems will ultimately realize precision CV medicine. Our T32 fills a unique niche in CV data science that is currently missing, both at UCLA and nationally. The UCLA Integrated Data Science Training in CV Medicine (iDISCOVER) Program draws upon faculty from the UCLA Schools of Medicine and Engineering, to establish a program for trainees committed to intensive data science applications in CV medicine. We have a substantiated track record in establishing training programs, as evidenced by our NIH Big Data to Knowledge (BD2K) Initiative, Heart BD2K Center at UCLA. Experience has enabled us to construct a T32 research program targeting the most pressing data science questions in CV medicine. Our two-year program will accept qualified students who have completed 1st year PhD training from Computer Science (CS), Bioinformatics (BI) or Bioengineering (BE); and eligible postdoc fellows from elite CV programs. We will train predoctoral students during their second and third year of PhD training, and postdoctoral fellows during their first and second year of their fellowship. Trainees will engage in advanced coursework within the specific focus areas: (i) omics phenotyping-supported outcome studies; (ii) machine learning-supported approaches in CV medicine; and (iii) information indexing and knowledgebase construction. Trainees will engage in CV clinical rotations to give them exposure to pressing CV data science questions. Trainees will be guided by a co-mentoring arrangement (1 CV mentor and 1 data science mentor). We have an outstanding group of 14 core faculty and 6 clinical supporting faculty members in the UCLA Schools of Medicine, Engineering and Life Sciences. Our faculty have established, vibrant and well-funded research programs with strong histories of guiding students to successful careers. Our program promotes the training of underrepresented minority groups, as demonstrated by training records of all faculty. Our iDISCOVER program has the institutional backing from Schools of Medicine, Engineering, Grad Division, Depts. of CS, BI, and BE, as well as Cardiology and Physiology. These elements ensure solid support for program implementation, advancement and success.
PROGRAM NARRATIVE The UCLA iDISCOVER Program aims to train and inspire a new generation of cardiovascular informaticians who will lead the development of innovative approaches that inform and enable precision cardiovascular medicine. iDISCOVER establishes a cross-campus, interdisciplinary environment to teach and mentor future scientists in cutting-edge computational and data science methods; and encourages novel, team science research opportu- nities for engineering graduate students to work with and learn from leading experts in cardiovascular medicine. iDISCOVER trainees are exposed to the breadth of contemporary biomedical informatics research, and are ulti- mately prepared to be productive, independent scientists that will participate in shaping the discipline of modern data science-driven cardiovascular research and healthcare.
