This talk will discuss how modern data mining techniques can be imported into statistical genetics. Most relevant models now invoke high-dimensional optimization. Penalization and set projection give sparsity. Separation of variables gives parallelization. Time permitting, these ideas will be illustrated by several examples: estimation of ethnic ancestry, genotype imputation via matrix completion, conversion of imputed genotypes into haplotypes, matrix completion discriminant analysis, estimation in the linear mixed model, iterative hard thresholding in GWAS, and sparse principal components analysis.
The U-M Department of Biostatistics is holding a two-day Symposium on Big Data, Human Health and Statistics on June 25-26 as the closing event of its summer institute on Transforming Analytical Learning in the Era of Big Data.
Scheduled speakers are:
- Susan Murphy, Statistics, Psychiatry, and Institute for Social Research, U-M
- Jeremy M.G. Taylor, Biostatistics, Radiation Oncology, U-M
- Goncalo Abecasis, Biostatistics, U-M
- Jenna Weins, Computer Science and Engineering, U-M
- Todd Mostak, MapD
- Rachel Schutt, News Corp., Columbia University
For a detailed agenda and to register, visit the symposium website. The registration deadline is June 15.
The U-M Department of Biostatistics is holding the Undergraduate Summer Institute in Biostatistics from June 1 – 26, 2015. The theme is “Transformational Analytical Learning in the Era of Big Data,” and applications are being accepted through March 15.
Description:
The field of big data science that intersects with public health and biomedicine is changing rapidly with datasets of enormous complexity and size being gathered in diverse areas including genomics, imaging, electronic health records, social media and environmental monitoring. The training of the next generation of quantitative scientists needs to change to meet the demands of the data. We define “Big Data” as datasets of enormous size and complexity (either in number of observations, and/or in the number/nature of predictors/outcomes). Classical theory, computation and intuition often fail for such irregular, sparse data sets of vast size. More training in data management, data storage, visualization, high dimensional statistics, optimization, causal methods, modeling sparse data and machine learning are needed to equip students to tackle these big data challenges. It is expected that the knowledge obtained from these massive heterogeneous data sources will inform prevention, screening, prognosis and treatment of human diseases and play a major role in biology, medicine and public health in the coming decade.
This full-time 4 week summer institute held in the Ann Arbor campus of the University of Michigan is targeted toward undergraduates who have an interest (or are susceptible to being interested) in the intersection of Big Data, Statistics, and Human Health. The institute is led by a distinguished group of faculty from the Department of Biostatistics at the University of Michigan School of Public Health (UMSPH) with additional outstanding faculty from Statistics and Electrical Engineering and Computer Science (EECS).
To apply:
Visit the institute’s “How to Apply” page.