20 May 2016
Ramy Arnaout
Beth Israel Deaconess Medical Center, Boston
We study B- and T-cell repertoires. One aspect we are interested in is repertoire diversity, since it is both clinically important and a key measure of immunological complexity. We would like to measure diversity in the entire individual but we are limited to measurements on samples. Even though our measurements are based on millions of cells — a "big" number — this distinction turns out to be a big issue. Inherent uncertainty in the number of B- and T-cell clones that will be missing from a blood or tissue sample by chance (the missing-species problem), inevitable sampling bias, and experimental noise make it difficult to extrapolate from sample diversity to overall diversity. Today I will talk about these pitfalls in theory and practice and outline a few solutions, including a modified maximum-likelihood method we developed that outputs the overall diversity of a repertoire from measurements on a sample.