Peter Visscher has asked me to forward this message to the list.

Dave Waddington


FIRST (PRELIMINARY) ANNOUNCEMENT

SURVIVAL ANALYSIS APPLIED TO ANIMAL GENETICS
AND EPIDEMIOLOGY

We are pleased to announce a postgraduate course in
"Survival Analysis applied to Animal Genetics and
Epidemiology" to be held at the Institute of Cell, Animal and
Population Biology (ICAPB) of the University of Edinburgh in
April 2000.

Teacher: Dr. Vincent.P. Ducrocq (INRA, Jouy en
Josas, France)
 
Dates: 10-14 April 2000
 
Location: ICAPB, University of Edinburgh, King's
Buildings, Edinburgh, Scotland
 
Course fee: These will depend on the total number of
participants. We aim to charge ~£50 per postgraduate
student, ~£100 per postdoc/academic, and ~£200 for
industry participants. Costs will include notes,
morning/afternoon coffees, lunches. The exact costs will be
set before the next announcement (January 2000)
 
A course description and more details will be placed on the
WWW in due course (http://helios.bto.ed.ac.uk/icapb/).

COURSE DESCRIPTION
 
Survival Analysis applied to Animal Genetics and
Epidemiology
Vincent P. Ducrocq
 
Objectives

Animal scientists (animal geneticists, epidemiologists,
economists, etc...) and biologists are often studying ways to
increase the average length of productive life of domestic or
experimental animals or to decrease the frequency of
involuntary cullings, and, in general, to understand variation
in longevity. In human medicine, there is great interest in
the genetics of longevity and age-at-onset of diseases.

Survival analysis represents a special field in statistics
because it deals with two types of data : "complete"
records, when the "failure time" of the individual is known
exactly, and "censored" records corresponding to the
"current" length of life when the individual is still alive at the
end of the study period. Sophisticated methods have been
developed in the biomedical world to better use all the
information available (from both uncensored and censored
records). The use of such methods in animal and human
genetics and epidemiology appears particularly promising.

The objective of the week is to present the basic
methodology of survival analysis. To facilitate the
understanding of the concepts involved and to stimulate the
use of these techniques in data analyses, practicals will be
included each day. These practicals will make use of the «
Survival Kit », (Ducrocq and Sölkner, 1998), a software
specifically developed for survival analysis in animal
genetics and epidemiology. In particular, it can be used to
study the influence of time-dependent effects (such as
disease occurrence or milk production in lactating
animals). Random (e.g., genetic) effects can be included
and their variance can be estimated. Some recent
applications will also be presented.

Program outline :
Day 1 : Introduction to survival analysis ; analysis of
homogeneous
populations
Day 2 : Parametric and non parametric regression models
Day 3 : Generalization (use of time-dependent covariates) ;
some
computational aspects
Day 4 : Frailty (mixed) models ; estimation of genetic
parameters
Day 5 : Applications in animal genetics and epidemiology :
some examples
 
Requirements :
The presentation of the methodology of survival analysis will
assume some knowledge about probability distributions
and basic statistics concepts. But the main requirement is
that the participants should have already some experience
with data analysis.
 
 
About the teacher
 
Vincent Ducrocq obtained his PhD in Animal Breeding (with
minors in Statistics and Computer Science) from Cornell
University ( USA). He is currently Senior Research
Scientist with INRA, and is located at the Station on Applied
and Quantitative Genetics (SGQA) of the Department of
Animal Genetics at Jouy-en-Josas, France. He is there
leader of the dairy cattle group, and he is also involved in
teaching. His research activities are related to genetic
analysis of secondary traits in dairy cattle. He is in charge
of the routine evaluation of type traits and longevity in dairy
cattle in France. He has given a course on survival
analysis earlier in Denmark, Germany, USA, The
Netherlands, Australia and France.
 
Some key papers on survival analysis are:
 
 
DUCROCQ V., QUAAS R.L., POLLAK E.J., CASELLA G.,
1988. Length of productive life of dairy cows. 1:
Justification of a Weibull model. J. Dairy Sci., 71, 3061-
3070.

DUCROCQ V., QUAAS R.L., POLLAK E.J., CASELLA G.,
1988. Length of productive life of dairy cows.2: Variance
component estimation and sire evaluation J. Dairy Sci., 71,
3071- 3079.

BEAUDEAU F., DUCROCQ V., FOURICHON C.,
SEEGERS H., 1995. Effect of disease on length of
productive life of French Holstein dairy cows assessed by
survival analysis. J. Dairy Sci., 78, 103- 107.

DUCROCQ V., CASELLA G., 1996. A Bayesian analysis of
mixed survival models. Genet. Sel. Evol, 28, 505-529.

ESSIOUX L, ABEL L AND BONAITI-PELLIE (1995). Genetic
epidemiology of breast cancer: interest of survival analysis
methods. Ann Hum Genet 59: 271-282.

GRÖHN Y.T., DUCROCQ V., HERTL J. A., 1997. Modelling
the effect of a disease on culling : an illustration of the use
of time dependent covariates in survival analysis. J. Dairy
Sci., 80, 1755-1766.

PETERSEN JH (1998). An additive frailty model for
correlated life times. Biometrics 54: 646-661.

YASHIN AI & IACHINE IA (1995). Genetic analysis of
durations: correlated frailty model applied to survival of
Danish twins. Genet Epid 12: 529-538.


Please express your interest in participating and registering
for this course by email to me ([log in to unmask]).
A formal announcement for registration (including payment)
will be made early next year. Preference will be given to
postgraduate students.

Peter Visscher
------- End of forwarded message -------
===============================
Peter Visscher
Institute of Ecology & Resource Management
University of Edinburgh
West Mains Road
Edinburgh EH9 3JG
Scotland, UK
tel. +44 131 535 4052
fax. +44 131 667 2601
email: [log in to unmask]
===============================


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%