SM&FT 2006

 

Outcome prediction in early-discovery breast cancer:  Hope, Hype and Physics.

 

 

 

 

Eitan Domany

 

 

Abstract

 

Considerable effort has been devoted during the recent five years to identify a gene expression signature that predicts outcome of early-discovery breast cancer. Different  groups used different cohorts of patients and different  DNA microarrays to produce short-lists of predictive genes,  and reported high success rates.

 

I will review some of this work, point out problematic aspects of it and present PAC-ranking, a method designed to estimate  the number of training samples needed to produce a robust  predictive gene list.