Friday, December 11, 2009
Statistical Mechanics of transport through Nano-Channels: Biology, Physics and Nanotechnology
Functioning of living cells requires selective transport of proteins and other molecular signals into and out of the cell, as well between various cell compartments. Much of such transport is conducted by nano-scale channels, which function without direct input of metabolic energy and without channel transitions from an ‘open’ to a ‘closed’ state during transport. Mechanisms of selectivity of such channels provide inspiration for bio-engineering applications, in particular for design of selective nano-molecular sensors and sieves.
Precise mechanisms of selective transport through such ‘always open’ biological and artificial nano-channels are still unknown. I will present a theoretical model to explain the mechanisms of selectivity of transport through such nano-channels. The theory provides a simple physical mechanism for selectivity based on the differences in the kinetics of transport through the channel between different molecular species. In particular, the theory explains how the channels can remain selective in the presence of vast amount of non-specific competition. The theoretical predictions have been verified in experiments and have lead to the creation of a prototype artificial molecular nano-filter.
Friday, October 9, 2009
Acquisition of Clinical Data
Albany Medical Center
An informal discussion of how one might acquire relevant data to support one's research. We will start off with defining clinical trial, then discuss on where to go to gather data and followed by designing a study. This will lead us into an example of an ongoing trial. We will also review how any research will be used in everyday clinical practice. Finally, how can clinicians and statisticians co-exist?
Friday, September 25, 2009
Mathematician and Statistician Working for Insurance Companies — New York Life
A presentation on the various professional activities that Mathematicians and Statisticians perform for insurance companies from policy formulation to investment strategies and opportunities for internships and permanent positions.
Friday, September 18, 2009
A dose-response trend test in long-term animal carcinogenicity experiments
This talk will cover the existing statistical methods of estimations and hypothesis testing in long-term animal carcinogenicity experiments. Especially I will focus on the poly-\(k\) test and its modifications for survival/sacrifice experiments or single sacrifice experiments. The poly-\(k\) test is a modified Cochran-Armitage trend test on tumor onset probabilities obtained by adjusting mortality of animals, where \(k\) must be selected a priori to perform the test.
Friday, September 11, 2009
Synchronization of stochastically coupled oscillators: dynamical phase transitions and large deviations theory (or birds and frogs)
Systems of oscillators coupled non-linearly (stochastically or not) are ubiquitous in nature and can explain many complex phenomena: coupled Josephson junction arrays, cardiac pacemaker cells, swarms or flocks of insects and birds, etc. They are known to have a non-trivial phase diagram, which includes chaotic, partially synchronized, and fully synchronized phases. A traditional model for this class of problems is the Kuramoto system of oscillators, which has been studied extensively for the last three decades.