Online Program

Building a Community of Competent Statistical Practitioners in Proteomics (304893)

Keywords: statistical bioinformatics, proteomics, open-source software, education, reproducible research

Statistical methods and software are key for reproducible research. This is certainly true in quantitative mass spectrometry-based proteomics, which must overcome many sources of bias and unwanted variation. This talk will illustrate our efforts of promoting sound statistical practice and reproducible research, and building a community of competent practitioners, in this relatively new and rapidly evolving field. First, we will present the general but flexible framework based on linear mixed effects models, and the associated open-source software, which reflects the needs of the experimentalists and maximizes the reproducibility of the results across various choices of data processing. Second, we will detail our efforts of building a community of competent practitioners of statistics, through a world-wide series of short courses intended for experimentalists and computational scientists alike. Finally, we will highlight the opportunities that these efforts offer for training a new generation of "multilingual" students with computational and statistical background, who can communicate with scientists across disciplinary barriers.