ENSPM 2021 Talk
Statistics and Geometry for biological systems
Distances are an essential component of modern multivariate statistics and bioinformatics.
One can do statistics in spaces of complex objects such as trees, networks, shapes and images.
However geometry is not enough as the real data are never uniformly distributed on latent manifolds but occur with varying densities which are hard to capture when the data are sparse.
Using prior information one can incorporate data and construct posterior distributions along nonlinear dimensions and provide meaningful approximations to complex data even in non-Euclidean settings.
I will provide examples of using both mathematical and computational tools to understand trajectories followed by the human microbiome and maybe even an understanding of how food ingredients are hared across the world.
This contains joint work with Lan Huong Nguyen, Elisabeth Purdom, Christof Seiler,
Nina Miolane, Claire Donnat and Kris Sankaran.
Susan Holmes (Stanford University, USA)
Statistician and professor at Stanford University, Holmes is noted for her work in applying nonparametric multivariate statistics, bootstrapping methods, and data visualization to biology. She uses computational statistics, in particular, nonparametric computer intensive methods such as the bootstrap and MCMC to draw inferences about many complex biological phenomena, interactions between the immune system and cancer, resilience and biomarker detection in the human microbiome and drug resistance in HIV.
She often uses and contributes to mathematics. One celebrated work is on the geometry of tree space, introducing distance between phylogenetic trees to make a Cat(0) complex. Received her PhD in 1985 from Université Montpellier II, served as a tenured research scientist at INRA for ten years, taught at MIT, Harvard, and was an associate professor of biometry at Cornell before moving to Stanford in 1998.
More info >