Erythrocyte isozymes, other polymorphisms, and the coefficient of kinship in northeastern Brazil

Typescript.Thesis (Ph. D.)--University of Hawaii, 1969.Includes bibliographies.1 v. (various pagings) illus., tablesApplication of zymogram development technique has become a powerful tool in genetics research. It provides easy detection of molecular variation in proteins, leads to the discovery of new polymorphisms, and significantly adds to the knowledgment of man's genetical makeup and population structure. This dissertation comprises a study of isozymes and other polymorphisms, followed by their application to the analysis of population structure. During the year 1962-63 a total of 1068 families was selected from a migrant population of northeastern Brazil. Red blood cells and sera were collected and have been kept in satisfactory condition for study of enzymes and proteins. The red cell enzymes glucose-6-phosphate dehydrogenase, phosphogluconate dehydrogenase, phosphoglucomutase and adenylate kinase were studied by starch gel electrophoresis. Except for G6PD whose gene frequencies depend on the proportion of Negro ancestry, no other association was detected when gene frequency was regressed on other variables such as social level, sex, age, longitude, latitude, and regional pathology (hepatosplenomegaly). Two families carrying an unusual variant in the G6PD system are described, accompanied by biochemical characterization of the new allele v and cross-check with other G6PD mutants. Once proved to be unique, the new all8le is provisionally designated as Gd Minas Gerais (MG), after the Brazilian state of the propositus. Three families carrying rare alleles at the PGMZ locus are also described. A linkage study between isozymes and other genetic markers was done, mainly stimulated by a recent report claiming a probable linkage between the phenylthiocarbamide (PTC) and the phosphoglucomutase (PGM1 ) loci. We studied twenty-eight families where one of the parents was doubly heterozygous, i.e., double and single backcrosses, and four intercross families. We did not detect linkage in our sample either by analyzing the recombination frequencies by sex, nor in the pooled sample. Most lod scores are negative, giving evidence against linkage, and close linkage (θ ≤ .2) may confidently be excluded. By using other segregating families we also tested linkage between hemoglobin and PGM, hemoglobin and Gc, hemoglobin and cholinesterases E1 and E2. None of the observed lod scores were suggestive of linkage. The isozyme, and other polymorphisms which were previously studied in the same sample, are used to bioassay the population structure of northeastern Brazil. The migration model of Malecot was selected because it estimates the most pertinent parameters ruling the structure of a population (migration, systematic pressure, and consanguinity) besides its many-sided applicability to any genetic system (phenotype), isonomy, metric data, and pedigree inbreeding. A computer program DISTAN draws random pairs of individuals, groups them into classes according to the distance between their birth-places, and estimates the coefficient of kinship for every distance class. Estimation of the parameters a (mean coefficient of kinship for individuals born at small distance), b (measure of systematic pressure), and £ (dimensionality of migration) are also given by the program DISTAN. According to Malecot's theory, if the distance between birth-place of individuals I-and J is d, then the coefficient of kinship is given by φ(d) = ae^-bd d^-c , and is defined as the probability that two alleles drawn at random, one from individual I and one from individual J, be identical by descendent from a common ancestor. The monotonic decrease with distance of inbreeding and kinship predicted by Malecot has been observed in northeastern Brazil for polymorphisms, metrics, surname concordance (isonomy), and pedigree inbreeding. Comparison of these indicators with each other and with Malecot's theory indicates that consanguineous marriage at a given distance is preferential, most conspicuously at large distances. The mean coefficient of inbreeding is estimated to be .0080, of which 70 per cent was ascertained through pedigrees. Interpretation of these results is discussed in terms of various concepts of population structure