http://swrc.ontoware.org/ontology#Article
Sampling errors in phylogeny
en
TAKAHATA Naoyuki
TAJIMA Fumio
尚之 高畑
The sampling variance of nucleotide diversity or branch length in a phylogenetic tree constructed by any distance method provides a criterion to judge whether a deduction or an inference made from data is statistically significant. However, computation of the sampling variance is usually tedious, particularly when the number of operational taxonomic units (OTUs) or DNA sequences is large, and must rely on computers. Recently, Nei and Jin (1989) have developed a computer algorithm, but it can be applied only to a simple substitution model. In this paper, we derive simple formulas for the minimum and maximum values of the sampling variance, which are independent of underlying substitution models. Application of these formulas demonstrates satisfactorily accurate estimates of the sampling variances and therefore their practical use.
Molecular Biology and Evolution
8
4
494-502
1991
Oxford University Press
07374038
© 2013 Society for Molecular Biology and Evolution
http://mbe.oxfordjournals.org/content/8/4/494.full.pdf+html?sid=9789779e-b1f9-4af0-98e1-badef09b08ba|Society for Molecular Biology and Evolution