BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Nanostatistics – Statistics for Nanoscopy
DTSTART:20170214T171500
DTEND:20170214T190000
DTSTAMP:20260407T101156Z
UID:658d9141bbe8f5197b011874d351a908b7d647a0df2b29b81c8b876b
CATEGORIES:Conferences - Seminars
DESCRIPTION:Prof. Axel Munk University of Goettingen and Max-Planck Instit
 ute for Biophysical Chemistry\nConventional light microscopes have been us
 ed for centuries for the study of small length scales down to approximatel
 y 250 nm. Images from such a microscope are typically blurred and noisy\, 
 and the measurement error in such images can often be well approximated by
  Gaussian or Poisson noise. In the past\, this approximation has been the 
 focus of a multitude of deconvolution techniques in imaging. However\, con
 ventional microscopes have an intrinsic physical limit of resolution. Alth
 ough this limit remained unchallenged for a century\, it was broken for th
 e first time in the 1990s with the advent of modern superresolution fluore
 scence microscopy techniques. Since then\, superresolution fluorescence mi
 croscopy has become an indispensable tool for studying the structure and d
 ynamics of living organisms\, recently acknowledged with the Nobel prize i
 n chemistry 2014. Current experimental advances go to the physical limits 
 of imaging\, where discrete quantum effects are predominant. Consequently\
 , the data is inherently of a non-Gaussian statistical nature\, and we arg
 ue that recent technological progress also challenges the long-standing Po
 isson assumption. Thus\, analysis and exploitation of the discrete physica
 l mechanisms of fluorescent molecules and light\, as well as their distrib
 utions in time and space\, have become necessary to achieve the highest re
 solution possible and to extract biologically relevant information. In th
 is talk we survey some modern fluorescence microscopy techniques from a st
 atistical modeling and analysis perspective. In the first part we  addres
 s  spatially adaptive multiscale deconvolution estimation and testing met
 hods for scanning type microscopy. We illustrate that such methods benefit
  from recent advances in large-scale computing\, mainly from convex optimi
 zation. In the second part of the talk we address challenges  of quantita
 tive biology which require more detailed models that delve into sub-Poisso
 n statistics. To this end we suggest a prototypical model for fluorophore 
 dynamics and use it to quantify the number of proteins in a spot.
LOCATION:CM 1 4 https://plan.epfl.ch/?room==CM%201%204
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR