From: ncbi-admin@ncbi.nlm.nih.gov on behalf of Richard Desper
[desper@ncbi.nlm.nih.gov]
Sent: Friday, May 18, 2001 2:17 PM
To: ncbi-seminar@ray.nlm.nih.gov; GIG-L@LIST.NIH.GOV
Subject: CBB seminar May 22

Room:  Building 45(Natcher), 6th floor conference Room South (ConfN6S)
Time:  Tuesday, May 22, 11:00 a.m.

Title: On the trail of the third gene for hereditary 
suscepitbility to breast cancer.

Speaker: Richard Desper

Abstract:

A significant proportion of familial breast cancers cannot be
explained by mutations in BRCA1 or BRCA2 genes.  We applied a
strategy using mathematical models on CGH data to identify
early somatic genetic deletions in tumor tissues, followed by
targeted linkage analysis.  This two-step process, using CGH
data to generate hypotheses for targeted linkage, had previously
been used to detect the locus for Peutz-Jeghers' cancer syndrome.

CGH, comparative genomic hybridization, is used to detect which
chromosomal regions witness over-duplication of oncogenes, or the loss
of tumor suppressor genes.  I will discuss the heterogeneous
nature of CGH data and show how simple modelling borrowed from
phylogenetics can be quite useful in providing a simple way to express
first-order dependencies in the data.  This modelling provides a
coherent way to consider the likely relative order of CGH events
and can point towards the discovery of tumor subclasses.  In particular,
it can be useful as a first step in hypothesis formulation, directing
the researcher towards specific regions in the genome which may be of
interest for targeted, closer analysis.  

These methods were used as part of a study of 61 breast tumors from 37
breast cancer families with no identified BRCA1 or BRCA2 mutations.  The
methods discussed, along with targted linkage, provided several lines of
evidence indicating that a third breast cancer gene is located on chromosome
13.