2020-03-04 MAR

(today's session fraught with agenda conflicts)

Preceding confluence meeting ends at 10am, an important BB seminar starts at 10:30. This agenda is therefore very tentative and most likely aspirational ... (i.e. it might be moved to next week). Pending topics:

PWA

What are they, how are they developed, give it a try creating a web worker, have a go at one of the toolboxes

Computational Statistics

How low can your computational statistics go? Let's try one example,

... code cross-tabulation of euclidean distances between iris flowers (https://episphere.github.io/qaqc/iris.csv) and determine exact average value

What about advanced procedures such as non-linear regression, how hard would it be to code them:

an example: https://github.com/jonasalmeida/fminsearch

AI explanation (Jeya)

Novelty Index and more ...

BB Seminar - 10:30-11:30

BIOSTATISTICS BRANCH SEMINAR SERIES PRESENTSBB seminar:Nilanjan Chatterjee, Ph.D.Bloomberg Distinguished ProfessorDepartment of Biostatistics, Bloomberg School of Public HealthDepartment of Oncology, School of MedicineJohns Hopkins University Date: Wednesday, March 4th, 2020Time: 10:30 am to 11:30 amLocation: 6E032/034 – Shady Grove/ 9609 Medical Center Drive Title: Next Generation Statistical Methods in the Post- Genome Wide Association Studies Abstract: In the early years of genome-wide association studies, data analysis primarily relied on fairly simplistic methods, such as running millions of univariate linear or logistic regressions, one for each genetic variants. Recently, however, as the sample sizes for some GWAS have become extremely large and various types of other genomic data have increasingly become available, analysis of such data has also become much more complex and statistically sophisticated. Because this field is mature and stable, there is tremendous opportunity to develop novel methods to address new types of questions using existing or anticipated data sources. In this talk, the speaker will try to make a case for the golden opportunities for statisticians by drawing examples from our recent work in the areas of estimation of genetic effect-size distribution, integrative analysis of GWAS and eQTL studies, modeling tumor heterogeneity and Mendelian randomization. The speaker will try to highlight gaps in these and other areas that remain fertile areas of future research.