COMPUTER PROGRAMS FOR GENETIC ANALYSIS OF QUANTITATIVE CHARACTERS
Shaw (1987; Evolution 41: 812-826 and 1991; Evolution 45: 143-151)
and others have presented a maximum-likelihood approach to quantitative
genetic analysis. This approach provides estimates of fixed effects
and components of variance for a polygenic model fitted to data on
phenotypic values for individuals whose genetic relationships are
known. I am now making available new versions of my programs to conduct
these analyses. Numerous improvements distinguish the current versions
from those I have distributed previously. The current programs have
the capability of handling multivariate data of diverse genetic designs,
spanning the range of standard mating schemes (offspring-parent,
nested, factorial, diallel, regardless of balance). The programs permit
tests of the hypothesis that any particular component (or set of
components) of variance or covariance equals zero. In addition, a
program that tests the hypothesis that two G matrices are the same
is also now available. Efficiency of the current programs (i.e.
requirements of computer memory and time) has been greatly improved
over that of my previous versions. Data input has been made more
flexible and diagnostics have been implemented in order to avoid
and/or clarify runtime errors. The programs are written in standard
Pascal and are distributed as source code. They have been run successfully
on Sun and DEC workstations, VAX and IBM mainframes, and a Cray Y-1
supercomputer. Developments in testing phase and planned for
distribution in the near future include 1) a program that accommodates
data taken on general pedigrees spanning multiple generations and 2)
a version that tests the hypothesis that the genetic correlation
between two traits equals one.
.pp
.ps +2
Programs and accompanying files can be obtained by sending a request
together with a single DOS-formatted diskette (either size) to the
address below. Alternatively, files are available in the IuBio
archive of biology software and data maintained by D.G. Gilbert.
This archive is accessible using standard anonymous ftp over the
Internet network to ftp.bio.indiana.edu. An overview of the IuBio
archive and general instructions are given in the file 'Readme'.
The ML quantitative genetics programs and associated files are
located in the /biology/quantgen/quercus subdirectory. Within it,
see the file Readme.qg for an outline of currently available files.
Please send me your name and address, including e-mail address, to
notify me that you have received programs. This work has been
supported by grants from the National Science Foundation (BSR 8817756,
BSR 8905808, and DIR 9112842).
.ce 5
Ruth G. Shaw
Department of Botany and Plant Sciences
University of California
Riverside, CA 92521
shaw at superb.ucr.edu
.sp
.ce4
after January 1, 1993
Department of Ecology, Evolution, and Behavior
University of Minnesota
St. Paul, MN 55108
