Non-linear feature selection in high-dimensional genomic data sets: Neyman Seminar

Seminar | October 30 | 4-5 p.m. | 1011 Evans Hall

 Chloé-Agathe Azencott, Mines ParisTech

 Department of Statistics

Many problems in genomics require the ability to identify relevant
features in data sets containing many more orders of magnitude than
samples. One such example is genome-wide association studies (GWAS), in
which hundreds of thousands of single nucleotide polymorphisms are
measured for orders of magnitude fewer samples.

This setup poses statistical and computational challenges, and for this
reason few approaches attempt to model non-additive effects. In my talk,
I will present methods we have developed to discover features that act
non-linearly on the outcome of interest.

References:
Lotfi Slim, Clément Chatelain, Chloé-Agathe Azencott, Jean-Philippe
Vert. Novel methods for epistasis detection in genome-wide association
studies BioRXiv, 2018.
Lotfi Slim, Clément Chatelain, Chloé-Agathe Azencott, and Jean-Philippe
Vert. kernelPSI: a post-selection inference framework for nonlinear
variable selection, ICML, 2019.

 Berkeley, CA 94720, 5106422781