Joint estimation of immigration and mating system parameters in gymnosperms using the EM algorithm

An EM algorithm procedure is presented for the maximum-likelihood joint estimation of immigration and mating system parameters of mixed-mating system models for gymnosperms. In addition to accommodating multiallelic and multilocus data in mature populations and pollen pools, the EM estimates are ins...

Full description

Saved in:
Bibliographic Details
Published in:Theoretical and applied genetics 1991-12, Vol.83 (2), p.137-140
Main Authors: Xie, C.Y. (Alberta Univ., Edmonton (Canada). Dept. of Forest Science), Yeh, F.C, Dancik, B.P, Strobeck, C
Format: Article
Language:English
Subjects:
ANALISIS ESTADISTICO
ANALYSE STATISTIQUE
Biological and medical sciences
Classical genetics, quantitative genetics, hybrids
DISTRIBUCION DE LA POBLACION
DISTRIBUTION DES POPULATIONS
EM algorithm
Fundamental and applied biological sciences. Psychology
Genetics of eukaryotes. Biological and molecular evolution
GIMNOSPERMAS
GYMNOSPERME
GYMNOSPERMS
Immigration
MATING SYSTEMS
Maximum-likelihood estimation
METHODE D'ACCOUPLEMENT
Methods, theories and miscellaneous
POPULATION DISTRIBUTION
SISTEMAS DE APAREAMIENTO
STATISTICAL ANALYSIS
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:An EM algorithm procedure is presented for the maximum-likelihood joint estimation of immigration and mating system parameters of mixed-mating system models for gymnosperms. In addition to accommodating multiallelic and multilocus data in mature populations and pollen pools, the EM estimates are insensitive to allelic frequency changes in foreign population and may approach closer to global maximum-likelihood estimates with each iteration, regardless of initial starting values. Estimates of rates of selfing (Ŝ), outcrossing (Ô), and immigration (Î) derived from the model are bounded strictly within the natural biological range (i.e., 0 ≤ Ô + Î ≤ 1; Ŝ + Ô + Î = 1).
ISSN:0040-5752
1432-2242
DOI:10.1007/BF00226241