BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Some Algorithmic Problems Arising in Systems Biology and Statistic
 al  Genetics
DTSTART:20090713T141500
DTSTAMP:20260407T101136Z
UID:6c8b6c6cc19a60512ebca0c9ff5b19881a127bff6730c0b1e51dd5e0
CATEGORIES:Conferences - Seminars
DESCRIPTION:Prof. Richard M. Karp\, University of California at Berkeley\n
 The quest to understand how living cells work and how genetic variation is
  related to disease requires the extraction of knowledge from large bodies
  of genomic and molecular data. This effort leads to large-scale combinato
 rial optimization problems. We present approaches to three such problems: 
 (1) Alignment of multiple genomes. One is given several closely related ge
 nomes in which pairs of highly similar segments called "anchors" are desig
 nated. The problem is to align the genomes so that\, to the greatest exten
 t possible\, matching anchors are aligned against each other. We formulate
  this problem as an implicit hitting set problem\, and develop and apply a
  generic approach to such problems. Joint work with Erick Moreno Centeno. 
 (2) Given a set of "query proteins" and a network in which the vertices re
 present proteins and the edges represent physical interactions between pro
 teins\,find the most compact connected subnetwork containing at least one 
 protein matching each query protein.Such subnetworks are likely to be func
 tional units in the regulation of cellular processes.We present solution m
 ethods based on dynamic programming\, integer programming and simple but e
 ffective heuristics. Joint work with colleagues at Tel-Aviv University. (3
 ) Finding associations between SNPs and disease. A SNP (single-nucleotide 
 polymorphism) is a site within a genome where two different nucleotides co
 mmonly occur. We discuss the construction of efficient sequential experime
 ntal designs to identify the SNPs most highly associated with a given dise
 ase.\nProf. Karp's webpage
LOCATION:CM 1 2 https://plan.epfl.ch/?room==CM%201%202
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR