Reproducibility in Personalized Medicine Research

During the past decade, one of the most notable transformations in science has been the availability of large and diverse sets of data. This trend has been accompanied by the increased use of machine learning and statistical techniques to classify patients and optimize treatments in precision medicine. In this context several concomitant factors can cause poor reproducibility levels, including unmeasured and heterogenous covariates’ distributions across studies, new technologies and ascertainment mechanisms. The focus of the workshop will be on statistical techniques and applications to understand and prevent the most important and common causes of lapses of reproducibility.

Sponsored by the Department of Biostatistics at the Harvard TH Chan School of Public Health.

Invited speakers:

Keith Baggerly, The University of Texas MD Anderson Cancer Center?
Bin Yu, University of California, Berkeley ?
David Madigan, Columbia University
Lorenzo Trippa, Harvard TH Chan School of Public Health and Dana-Farber Cancer Institute
Levi Waldron, City University of New York

Click here for agenda.

Student Poster Competition: Prizes (1st, 2nd, 3rd posters $500 each)

We invite students to present their work on novel statistical methods and computational approches. A broad range of contributions, from theoretical statistics to applications in a variety of biomedical fields will be evaluated. Students with ongoing projects in robust statistics, reproducibility and replicability are particularly encouraged to participate.

To submit your poster for consideration, please fill out the form at http://goo.gl/forms/WhdgrqDmTSkgHDxe2.