FluoRender is a tool developed at the SCI Institute for visualizing and analyzing volume data acquired from fluorescence microscopy. Since its public release in 2009, it has aided many biomedical researchers, especially neurobiologists. In this talk, we examine several recent-year publications in biology journals. The research in these publications all used FluoRender. The publications are grouped into three categories: general data visualization, time-dependent data visualization, and insect brain anatomy. For each case study, I will present the biological questions that the researchers endeavored to answer, their approaches, and results. I will highlight each case with a plenty of illustrations, figures, and videos from the original work, adding my own interpretations and sometimes demo videos of our own. These case studies not only show us how our tools have been used but also provide insights for future work.
Posted by: Nathan Galli
Reproducing Kernel Hilbert Spaces (RKHS) appeared in a theoretical paper (Aronszajn 1950), but their use in applied nonparametric regression, statistical model building, machine learning and classification had to wait for modern computational facilities. We review RKHS and then Analysis of Variance (ANOVA) decompositions of functions of several variables in tensor products of RKHS.
We review Distance Correlation, which is a completely nonparametric approach for examining correlation between essentially arbitrary clusters of random variables, based on samples of pairwise distances. We marry these tools to examine how lifestyle and other variables in the Beaver Dam Eye Study correlate with mortality as it runs in families.