Project Summary/Abstract Biomedical research is evolving rapidly as new high-throughput technologies, producing massive quantities of data, open new frontiers in data-driven predictive capabilities, boosting the applicability of complex data analysis and machine learning algorithms in areas ranging from personalized medicine and drug discovery to structural biology and microbial ecology. Simultaneously, high standards for ethical production, use, and open access to data have become essential. This new application seeks support for the Predoctoral Training Program in Bioinformatics and Computational Biology at Boston University which will train young scientists to become leaders in this transformational era. Ten predoctoral training slots per year are requested to fund five trainees each in years one and two. The curriculum includes a strong foundation in biological domain knowledge, advanced methodologies in the quantitative sciences (computing, mathematics, and statistics), an emphasis on reproducibility habits, awareness of algorithmic and racial bias in research, and extensive opportunities for developing scientific communications skills. Program features include: 1) three lab rotations, including the Wet-Lab Experience, which introduces new trainees to high-throughput experimental methods, 2) the Challenge Project, for first-year team research on open-ended, data-intensive biological problems, with an emphasis on rigor and reproducibility, 3) Programming workshops for computational skills development, 4) Algorithmic Bias workshops for recognizing biases in data collection and use, 5) a research-in-progress Student Seminar, 6) the annual International Workshop in Bioinformatics and Systems Biology, undertaken jointly with partner programs in Japan and Germany, 7) the annual Student-Organized Symposium, 8) a teaching requirement, and 9) the annual Program Retreat. Thirty-six faculty mentors from 14 departments, each with a strong quantitative component to their research, offer a wide range of interdisciplinary expertise in experimental, mathematical and computational approaches. All use rigorous and reproducible methods in their research, take an active role in Program activities, are committed to active mentoring, and will have taken mentor training by the start of Program funding. Three co-PDs, who have worked together for over a decade, will provide leadership. Each brings training strengths and administrative experience. An Executive Committee, including five additional faculty will oversee the Program, with direct input from the trainees through a Student Advisory Council. The Program offers extensive career development activities, multipronged avenues for student engagement, and a multidimensional scheme for ongoing evaluation.

Project Narrative This Program will train young scientists to become leaders in the use of complex data analysis and machine learning algorithms to transform the massive data sets generated by high-throughput experimental technologies into knowledge and action in areas ranging from personalized medicine and drug discovery to structural biology and microbial ecology. They will also learn how to recognize and ameliorate algorithmic and racial bias in the generation and application of data so that the benefits derived from improvements in public health and the treatment of human disease are distributed equitably.