Technion, IEM faculty - Statistics Seminar
 
Speaker: Yuval Nov, University of Haifa.
 
Title: Protein engineering by means of statistical analysis
 
Date: 20/11/2011
 
Time: 14:30
 
Place: Bloomfield-527
 
Abstract:  <http://ie.technion.ac.il/seminar_files/1320746163_Nov.pdf>
 
Or read this:
 
The talk reports advances in two statistics-based methods for protein
engineering. Both methods are aimed to save laboratory time, resources,
and labor.
 
The first method's goal is to combine optimally beneficial mutations of a
protein. This method is based on a stochastic sequence-activity model,
devised specifically for proteins, whose parameters are estimated from
experimental data, and which allows prediction of highly-active protein
variants. Two empirical applications of this method are discussed.
 
The second method is concerned with new criteria for determining the
sample size in a protein engineering technique called saturation
mutagenesis. By using a criterion of this type, one may reduce
significantly the sample size while compromising minimally the quality of
the best variant discovered. The method is also used to compare the
efficiency of four randomization schemes, which underlie the saturation
mutagenesis protocol.
 
---------------------------------------------------------
Technion Math. Net (TECHMATH)
Editor: Michael Cwikel   <techm@math.technion.ac.il> 
Announcement from:  <ynardi@ie.technion.ac.il>