Showing papers by "Robert R. Sokal published in 1998"

Local spatial autocorrelation in biological variables

Abstract: Spatial autocorrelation (SA) methods have recently been extended to include the detection of local spatial autocorrelation at individual sampling stations. We review the formulas for these statistics and report on the results of an extensive population-genetic simulation study we have published elsewhere to test the applicability of these methods in spatially distributed biological data. We find that most biological variables exhibit global SA, and that in such cases the methods proposed for testing the significance of local SA coefficients reject the null hypothesis excessively. When global SA is absent, permutational methods for testing significance yield reliable results. Although standard errors have been published for the local SA coefficients, their employment using an asymptotically normal approach leads to unreliable results; permutational methods are preferred. In addition to significance tests of suspected non-stationary localities, we can use these methods in an exploratory manner to find and identify hotspots (places with positive local SA) and coldspots (negative local SA) in a dataset. We illustrate the application of these methods in three biological examples from plant population biology, ecology and population genetics. The examples range from the study of single variables to the joint analysis of several variables and can lead to successful demographic and evolutionary inferences about the populations studied.

Spatial genetic structure of human populations in japan

Abstract: We studied spatial patterns for 24 allele frequencies representing 15 systems (blood antigens, enzymes, serum proteins, color blindness, and cerumen) in Japan. The total number of samples over all systems and localities is 1125. We investigated patterns of genetic variation graphically as interpolated allele frequency surfaces, as one-dimensional and directional correlograms, and by testing for the direction of maximal genetic autocorrelation. We examined the allele frequency surfaces by various techniques of spatial autocorrelation analysis and found 13 allele frequency surfaces from 9 genetic systems exhibiting significant spatial patterns. Several surfaces have clinal patterns along the major axis of the Japanese archipelago; others tend toward a maximum or minimum in south-central Honshu. Yet other allele frequencies show long-distance differentiation or patchiness. We discovered seven areas of rapid genetic change by using the wombling method. These areas largely reflect maritime and montane barriers, and some are associated with dialectal boundaries in these populations. The observed patterns support the hybridization or dual structure hypothesis for the peopling of Japan.

Factors affecting population variation in eastern Adriatic isolates (Croatia).

Abstract: Inhabitants of the Croatian islands of Brac, Hvar, Korcula, and the Peljesac Peninsula have been the subject of extensive previous studies of local population differentiation. Most of these studies used biological and ecological variables, but some also considered historical and sociological factors. In this study we use genetic, morphological, kinship, and language distance data, collected for individuals from 26 rural communities on the islands of Brac, Hvar, Korcula, and the Peljesac Peninsula in the Adriatic, to further explore the interaction of historical, sociological, and biological factors in small populations and to test the significance of some of these proposed causes. First, we use matrix correlation methods to evaluate the relationships among different types of distance measures. The specific measures of genetic distance used here do not correlate well with other measures of population distance, and it appears that for the studied genetic systems the populations are not strongly differentiated. As expected, kinship and language distances have a high degree of association. Morphological differences among populations seem to be more closely tied to kinship distances than to genetic distances. This may result from modification of some morphological features by environmental rather than genetic factors, or it may be attributed to extensive, selective, nonrandom emigration of the population during the first decade of the twentieth century. In the second part of our analysis we use matrix correlation methods to evaluate and possibly identify the external factors that have contributed to the population differences. Specifically, we use design matrices to test hypotheses that population differences can be explained by one of the following factors: geographic isolation on the islands and peninsula, distance from the mainland, geographic barriers within the islands and peninsula, and the historical factors that differentially affected the three islands and the peninsula. Most of these design matrices reflect geographic distances ; although correlations between morphological variables and simple geographic distance between localities were not significant, correlations between these localities and a design matrix incorporating geographic distance along with geographic barriers, such as bodies of water and mountain ranges, are particularly important for explaining distances among kin. Design matrices provide an important tool for quantifying the relationship between historical and geographic factors, and measures of population distance.