We examined mitochondrial DNA (mtDNA) variation in six Mennonite communities from Kansas (Goessel, Lone Tree, Garden View, Meridian, and Garden City) and Nebraska (Henderson) to determine their genetic structure and its relationship to population history. Mitochondrial DNA haplogroup and haplotype information were obtained from blood samples from 118 individuals. Molecular genetic variation was analyzed using diversity measures, neutrality test statistics, spatial analysis of molecular variance (SAMOVA), and multidimensional scaling plots. The Mennonite samples exhibited eight western European mtDNA haplogroups: H, HV0, I, J, K, T, U, and X. Comparable to other populations of European descent, haplogroup H was the most frequent in all six communities and ranged from 35% in Lone Tree to 75% in Old Order Mennonites from Garden City. Fifty-eight different mtDNA haplotypes were found in these groups with only one shared among all six populations. Haplotype diversities varied from 0.81 in Goessel to 0.96 in Henderson and Garden View. Multivariate statistical analysis of these populations indicates that these Anabaptist communities formed new congregations by fissioning along familial lines. Population subdivision of these communities into congregations supports previously documented patterns of fission-fusion. These haploid molecular data provide a more accurate reflection of biological relationships between midwestern Mennonite communities than evidence based on classical genetic markers.

The serotonin transporter promoter length polymorphism (5-hydroxytryptamine transporter length polymorphism; 5-HTTLPR) has long been implicated in autism and other psychiatric disorders. The use of selective serotonin reuptake inhibitors (SSRIs) has a positive effect on treating some symptoms of autism. The effects of these drugs vary in individuals because of the presence of the S or L allele of 5-HTTLPR. Studies performed on various autistic populations have found different allele frequencies for the L and S alleles. Allele frequencies and genotypes of the South African autistic populations (African, mixed, and Caucasian) were compared with matching South African ethnic control populations. The *S/*S genotype was found to be highly significantly associated with all the South African autistic ethnic populations. In the South African African population the *S/*S genotype was present in 7 (33%) of the autistic individuals but in none of the control subjects, yielding infinitely large odds of developing autism. The odds of developing autism with the *S/*S genotype compared to the *L/*L genotype increased 10.15-fold in the South African mixed group and 2.74-fold in the South African Caucasian population. The allele frequency of the South African autistic population was also compared with studies of other autistic populations around the world, and highly significant differences were found with the Japanese, Korean, and Indian population groups. The difference was not significant for the French, German, Israeli, Portuguese, and American groups. This is the first South African study of autistic individuals of different ethnic backgrounds that shows significant differences in allele and genotype frequencies of 5-HTTLPR. The results of this study open new avenues for investigating the role of transmission of the L and S alleles in families with autism in South Africa.

The biodemography of isolated populations with a subsistence economy is of special interest because the conditions to which the populations have been subjected are similar to those present over a large part of the history of our species, and hence the conclusions drawn can be extrapolated to other similar populations. In the present work, we used two recently employed techniques (self-organizing maps and Monmonier's algorithm) to study the genetic structure of a small rural region: Fuentes Carrionas (province of Palencia, Spain). The study was based on information contained in marriage registers and family reconstructions for the period 1880–1979. We calculated the coefficients of relationships (Hedrick) among the parishes (by isonymy and parent-offspring matrix) and studied their relationship with geographic distances (Mantel). Then, using Monmonier's algorithm, we examined the genetic barriers. Finally, by applying self-organizing maps, we studied the distribution of surnames. The correlation between genetic distance matrices estimated through surnames and gene flow was high (p < 0.01), and geographic distance was found to be significantly related to both surnames and gene flow (p < 0.01); geographic distance therefore seems to be the main factor of isolation. However, the genetic barriers revealed a region divided in two, and the surname distribution displayed an identical division: 29.56% of the surnames appeared almost exclusively in the northeastern half and 21.91% appeared only in the southwest. Both techniques afforded coherent and complementary results. Their combined use allows a detailed study to be made. Although geographic distance was the strongest determining factor, we detected others factors (orographic and socioeconomic) that seem to have left their mark on the genetic structure of Fuentes Carrionas.

The statement that pairs of individuals from different populations are often more genetically similar than pairs from the same population is a widespread idea inside and outside the scientific community. Witherspoon et al. [“Genetic similarities within and between human populations,” Genetics 176:351–359 (2007)] proposed an index called the dissimilarity fraction (ω;) to access in a quantitative way the validity of this statement for genetic systems. Witherspoon demonstrated that, as the number of loci increases, ω; decreases to a point where, when enough sampling is available, the statement is false. In this study, we applied the dissimilarity fraction to Howells's craniometric database to establish whether or not similar results are obtained for cranial morphological traits. Although in genetic studies thousands of loci are available, Howells's database provides no more than 55 metric traits, making the contribution of each variable important. To cope with this limitation, we developed a routine that takes this effect into consideration when calculating ω;. Contrary to what was observed for the genetic data, our results show that cranial morphology asymptotically approaches a mean ω; of 0.3 and therefore supports the initial statement—that is, that individuals from the same geographic region do not form clear and discrete clusters—further questioning the idea of the existence of discrete biological clusters in the human species. Finally, by assuming that cranial morphology is under an additive polygenetic model, we can say that the population history signal of human craniometric traits presents the same resolution as a neutral genetic system dependent on no more than 20 loci.

The polymorphism of short tandem repeats (STRs) is commonly used for human identity testing. Many commercial kits are available for this purpose, in which multiplex PCR for 10–16 STRs is usually used. For optimal results, the kits recommend rather limited conditions, which not all forensic samples can satisfy. We increased the efficiency of several commercial kits by modifying the components and reaction parameters (primer concentration and cycle number, and annealing and extension time). To simulate low copy number, we used 1:10 and 1:100 diluted samples compared to the recommended concentration. A change in cycle number, annealing and extension time, and primer concentration showed various results. Fine-tuning of PCR conditions by combining the described changes, decreasing the primer concentration, and increasing the annealing and extension time together with increasing the cycling number dramatically increased the efficiency of the reaction, even in low-copy-number simulated samples. Detailed information for several kit components or kits with different components targeting difficult sample conditions, including low copy number, would be of great help in forensics.