Premium
Kinship by isonymy and by gene frequencies: A comparison of population structures at different hierarchical population levels
Author(s) -
Paoli Giorgio,
Franceschi Marcello G.,
Taglioli Luca
Publication year - 1996
Publication title -
american journal of human biology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.559
H-Index - 81
eISSN - 1520-6300
pISSN - 1042-0533
DOI - 10.1002/(sici)1520-6300(1996)8:4<445::aid-ajhb4>3.0.co;2-x
Subject(s) - geographical distance , kinship , genetic distance , isolation by distance , population , distance matrices in phylogeny , geography , statistics , demography , genetic structure , mathematics , genetic diversity , sociology , anthropology , combinatorics
A comparison of population structures based on isonymy and on gene frequencies (ABO, Rh, Kell) was conducted for a sample of 28,205 individuals residing in three different provinces (Lucca, Massa Carrara, La Spezia) in northwest Italy, on the basis of both chronological and spatial subgroupings. Relationships between and within population subsamples were measured by means of kinship coefficients. The aim of this study was focused on kinship decay with geographic distance, associated with the great difference in location and variability between isonymic and genetic data. The analysis was carried out by R st statistics and regression analysis to test the fit of the isolation by distance models. Further, the R matrices were converted into a distance measure, and Mantel's permutation test was used to assess the correlation across isonymy, genetic, and geographic matrices. Whereas estimates of R st and isolation by distance parameters obtained from genetic and surname data pointed to a roughly comparable basic pattern of spatial differentiation in both chronological periods, the absolute values differ substantially. Both R st and a isolation by distance parameters estimated from genetic data were higher than those from surnames, indicating greater local isolation by genetic analysis. The standard errors of b obtained from surname data were much smaller than those computed from genetic data, indicating that the kinship by isonymy coefficients fit Malècot's model better than the kinship coefficients estimated by the genetic data. Squared correlation coefficients among geographic, surname, and genetic distance matrices supported the above interpretations. The strong localization of surnames, the different level of variability in surname and gene frequency data, and random variations (due to the number of alleles considered) seem to be the main reasons for the observed differences between the two data sets. © 1996 Wiley‐Liss, Inc.