Interesting-looking AAPA 2013 abstracts

Program here (pdf).
Natural selection acts to maintain diversity between Out of Africa and sub-Saharan African populations in genes related to neurological processes and brain development. JASON A. HODGSON1,5, ALI AL-MEERI2, CONNIE J. MULLIGAN3 and RYAN L. RAAUM4,5. 1Anthropology, New York University, 2Biochemistry and Molecular Biology, Sana'a University, Yemen, 3Anthropology, University of Florida, 4Anthropology, Lehman College and The Graduate Center CUNY, 5-, The New York Consortium in Evolutionary Primatology.

The Yemeni and Mozabite are closely related Out of Africa (OOA) populations from the Arabian Peninsula and North Africa respectively, while the Maasai are a sub-Saharan African (SSA) population. Using genome-wide SNP data (publicly available for the Mozabite and Maasai, and collected here for the Yemeni) we show the Yemeni to have ~7% and the Mozabite to have ~26% recent sub-Saharan admixture, while the Maasai have ~27% Middle Eastern admixture. We use an adaptation of the locus specific branch length method to look for the effects of natural selection on alleles introduced to the three populations through admixture. We specifically look for 1) the adaptive introgression of alleles from SSA into the Yemeni and Mozabite, 2) the adaptive introgression of alleles from OOA into the Maasai, 3) purifying selection of SSA alleles out of the Yemeni and Mozabite, and 4) purifying selection of OOA alleles out of the Maasai. We found correspondence in patterns of adaptive introgression and purifying selection between the populations for 18 genomic loci, all of which contain protein-coding genes. The correspondence in signatures of selection between three independent populations is strong evidence for natural selection, rather than the false positive signals common in genome-wide scans of selection. Strikingly, of the regions where purifying selection is acting to maintain diversity between the Out of Africa and sub-Saharan African populations, eight out of twelve genes with known ontologies are involved in neurological processes or brain development. A binomial test found this enrichment to be significant. This research was partially supported by NSF grant BCS-0518530.

Modern human phenotypic variation: Exploring patterns of differentiation within and between continents. MARK HUBBE1,2, DANILO V. BERNARDO3, TATIANA F. ALMEIDA4, TSUHENIKO HANIHARA5 and KATERINA HANIHARA6. 1Department of Anthropology, The Ohio State University, 2Instituto de Investigaciones Arqueológicas y Museo, Universidad Católica del Norte, 3Departamento de Genética e Biologia Evolutiva, Universidade de São Paulo, 4Unidade de Genética Médica, Faculdade de Medicina da Universidade de São Paulo, 5Department of Anatomy, Kitasato University School of Medicine, 6Department of Early Prehistory and Quaternary Ecology, Senckenberg Center for Human Evolution and Paleoecology, Eberhard Karls Universität Tübingen.

Current consensus is that modern human cranial phenotypic variation is a result of isolation-by-distance and neutral evolutionary processes, with natural selection acting most notably in some anatomical regions of populations living in extreme environmental conditions. Under this model, there is an underlying assumption that the rate of morphological differentiation in the past was uniform across the planet. The goal of this study is to test this assumption by comparing the morphological differentiation of human groups between and within geographic regions. We analyzed a large sample of male individuals from 135 human series. Craniometric variation was assessed through Fst estimates calculated from 33 linear measurements for each pair of series. Series were grouped into 15 geographic regions and the average Fst values within regions were then compared with the average linear geographic distances between series. Our results show a very strong linear correlation between average Fst values and geographic distances within each geographic region. However, between-regions analyses generally show lower correlations with distances, suggesting that geographic distance is not a good predictor for the Fst values between continents. These preliminary results indicate that the processes associated with the morphological differentiation within and between continents may have differed considerably, and probably a more significant influence of bottlenecks and natural selection. With the exception of the differences between North America and the Old World, isolation by distance may not be a good predictor of morphological differentiation between groups located in different continents. Alternative models should be taken into account to explain this pattern.

Measuring constraints on selection in human and chimpanzee life histories. JAMES H. JONES. Department of Anthropology, Stanford University.

An organism's life history is comprised of its age or stage-specific vital rates and trade-offs that bind them. Fitness in structured populations is given by the renewal equation and perturbation analysis of the renewal equation can be used to measure the force of selection on vital rates. While perturbation analysis provides information about the direction of selection, it is not informative about constraints that are central to life history theory. We derive a simple means of measuring the constraints on structured life histories. Assuming stabilizing selection, we use fitness elasticities to measure constraints on selection on the life cycle. The ratio of the fitness elasticities of two traits measures the slope of the constraint curve that binds them at equilibrium. We apply this method to understanding the differing constraints between human and chimpanzee life histories, focusing specifically on the trade-off between adult fertility and pre-reproductive survivorship. The shapes of the constraint curves linking fertility and pre-reproductive survival are quite different between the two species. Human curves show a pronounced U-shape, while the chimpanzee curves are more monotonic and shallow. This result suggests substantial costs associated with both very early and late reproduction in humans. Overall, we find that a unit of investment in fertility among humans must yield minimally five times the benefit of a unit of investment in infant survival. The switch from investment in infant survivorship to investment in further fertility is strongly influenced by the capacity of mothers' continued investment to contribute to the survival of offspring.

Using admixture mapping to identify genetic linkages with variation in human facial shape. DENISE K. LIBERTON1, PETER CLAES2, SANDRA BELEZA3, GREG BARSH3,4, HUA TANG3 and MARK D. SHRIVER1. 1Department of Anthropology, Pennsylvania State University, 2Centre for the Processing of Speech & Images, Katholic University of Leuven, 3Department of Genetics, Stanford University, 4HudsonAlpha Institute.

Using a dataset of 594 individuals from three populations with genetic ancestry contributions primarily from European and West African parental populations, we tested 72 SNPs in 45 selection-nominated candidate genes that are known to be involved in Mendelian craniofacial dysmorphologies as well as 68 ancestry informative markers with no known involvement with craniofacial variation for association with normal variation in facial shape. A dense mesh of over 7,000 high-density landmarks was placed on three-dimensional images of adult faces taken with the 3dMDface imaging system. Principal Component (PC) scores were then calculated as a representation of normal morphological variation across the surface of a face. A subset of the top PCs that showed significant associations with West African genetic ancestry was selected for the association study.

The program ADMIXMAP was used to test for admixture linkages between these 140 SNPs and facial shape variation. Sex, height, weight and BMI were included as covariates to control for the effects of body size on face shape morphology. After applying a correction for multiple testing, four SNPs were found to be significantly associated with facial variation. Two of these SNPs were located in craniofacial candidate genes, suggesting a role for these genes in determining normal facial variation. Interestingly, two of the AIMs, which are not located in candidate genes, were also significant. This suggests that these SNPs may be in admixture linkage disequilibrium with variants in the region that may have a previously unknown role with craniofacial variation. This study was funded by the National Institute for Justice, Grant 2008-DN-BX-K125; National Science Foundation, DDIG #0851815, Wenner-Gren Foundation Dissertation Fieldwork Grant #7967.

Is thicker better? Testing adaptation hypotheses for cranial vault thickness. HANNAH E. MARSH. Anthropology, University of Iowa.

Several adaptations have been proposed to account for variations in cranial vault thickness between populations of recent Homo sapiens, and between Homo species, including protection from interpersonal violence, and masticatory strain. The protection hypothesis states that codified interpersonal violence, in the form of ritualized, face-to-face dispute settlement, would result in increased bone thickness on the anterior vault. This hypothesis is mainly applied to Native Australians. The strain hypothesis states that vault thickness would co-vary with the intensity of masticatory forces, based on the placement and orientation of muscle attachments on the vault. In this study, I compare thickness of the frontal bone versus the parietal bones for 9 populations of recent H. sapiens to assess whether Australians have uniquely greater frontal thickness. Also, I will compare patterns of thicker and thinner bone with patterns of vault strain to determine any relationship of thickening to buttress against strain. I find that Australians are not the only population with thicker frontals than parietals, which does not support an adaptive interpretation specific to Australians. When comparing locations of strain to thickened vault regions, 26% of the sample had thickening in the midsagittal plane, the plane along which nuchal muscle strain radiates, while masticatory strain registers in the coronal plane, along the coronal suture, a location that does not seem to show increased thickness in this sample. These hypotheses are refuted. For cranial vault thickness, “thick” versus “thin” is an over-simplification. There are complex patterns of variation within individuals and the entire species. This research was funded by a Sigma Xi Grant-In-Aid of Research, and by the University of Iowa Graduate College and Athropology Department.

A man’s face reveals his body height: A GMM approach to ontogenetic and static allometry. KATRIN SCHAEFER1, SONJA WINDHAGER1, DENNIS E. SLICE2 and PHILIPP MITTEROECKER3. 1Department of Anthropology, University of Vienna, Austria, 2Department of Scientific Computing, Florida State University, USA, 3Department of Theoretical Biology, University of Vienna, Austria.

A large number of studies investigated which facial features lead to dominance attributions and other social inference, whereas separate studies assessed the role of body height in sexual selection. Facial allometry might link these approaches, but has received surprisingly little attention. To start with, we predict the presence of size cues in the adult face and relate them to the postnatal growth pattern. Our sample comprises 44 frontal photographs of Austrian boys (6–11 years) and men (17–33 years) together with their body height. On each photo 35 anatomical landmarks and 34 semilandmarks were digitized.

We estimated allometry via regressions of facial shape on log centroid size and body height in the full ontogenetic sample as well as in the subsample of adults only. The ontogenetic regressions were very similar, reflecting the high correlation between facial size and body height during ontogeny (r= 0.83). The static regression pattern of adult facial shape on body height resembled the ontogenetic pattern with an overall elongation of the mid- and lower face, a relative decrease in eye size as well as a thickening and lowering of the eyebrows. By contrast, the relationship between adult facial shape and facial size was less pronounced and not statistically significant.

These results show that there are substantial shape cues to body height in the adult male face (with taller men having the more mature features). This might operate as confound in studies of facial masculinity, leadership, and mate preferences.

Neolithic human mitochondrial haplogroup H genomes and the genetic origins of Europeans. WOLFGANG HAAK1, PAUL BROTHERTON1,10, JENNIFER TEMPLETON1, GUIDO BRANDT2, JULIEN SOUBRIER1, CHRISTINA J. ADLER1,11, STEPHEN M. RICHARDS1, CLIO SI. DER SARKISSIAN1,12, ROBERT GANSLMEIER3, SUSANNE FRIEDERICH3, VEIT DRESELY3, MANNIS VAN OVEN4, JONAS KORLACH6, SIMON YW. HO7, LLUIS QUINTANA-MURCI8, DORON M. BEHAR9, HARALD MELLER3, KURT W. ALT2, ALAN COOPER1 and THE GENOGRAPHIC CONSORTIUM13. 1Australian Centre for Ancient DNA, University of Adelaide, 2Institut fuer Anthropologie, Johannes-Gutenberg-Universitaet Mainz, 3State Museum for Prehistory Halle, State Office for Heritage Management and Archaeology Saxony-Anhalt, 4Department of Forensic Molecular Biology, University Medical Center Rotterdam, 5SA Pathology, SA Health, 6Pacific Biosciences, Pacific Biosciences, 7School of Biological Sciences, The University of Sydney, 8Institut Pasteur, Institut Pasteur, 9Rambam Medical Center, Rambam Medical Center, 10Archaeogenetics Research Group, University of Huddersfield, 11Institute of Dental Research, The University of Sydney, 12Centre for Geogenetics, Natural History Museum of Denmark, 13The Genographic Project, National Geographic Society.

Haplogroup (hg) H dominates present-day Western European mitochondrial (mt) DNA variability (>40%), yet was less prevalent amongst early Neolithic farmers (~19%) and virtually absent in Mesolithic hunter-gatherers. To investigate this haplogroup’s significance in the maternal population history of Europeans we employed novel techniques such as DNA immortalization and hybridization-enrichment to sequence 39 hg H mt genomes from ancient human remains across a transect through time in Neolithic Central Europe.

The results of our population genetic analyses reveal that the current patterns of diversity and distribution of hg H were largely established during the Mid-Neolithic, but with substantial genetic contributions from subsequent pan-European cultures such as the Bell Beakers, which expanded out of Iberia in the Late Neolithic (~2800 BC). Using a strict diachronic approach allowed us to reconcile ‘real-time’ genetic data from the most common European mtDNA hg with cultural changes that took place between the Early Neolithic (~5450 BC) and Bronze Age (~2200 BC) in Central Europe. This revealed the Late Neolithic (2800-2200 BC) as a dynamic period that profoundly shaped the genetic landscape of modern-day Europeans.

Furthermore, linking ancient hg H genome sequences to specific points in time by using radiocarbon dates as tip calibrations allowed us to reconstruct a precise lineage history of hg H and to calculate a mutation rate 45% higher than traditional estimates based on the human/chimp split.


  1. @n/a

    Hey, just found this blog through Google. I've been thinking about getting my genome sequenced. Could you recommend some reputable services?

    You seem to know a lot about this stuff, so I figured you'd also know which companies are reliable. I've heard stories about less than stellar outfits delivering questionable results. I'd like to avoid those kinds of guys.

    Any help would be appreciated. Btw, science isn't my strong suit, so please explain it to me like I'm five. :-)


  2. 23andMe would probably be a good place to start.

    I don't know what your exact goals are but 23andMe is cheap and user friendly and will give you most of the currently useful information about your genome. Of the major companies, 23andme has the best ancestry analysis. If you're interested in ancestry/genealogy, they also have a very large "Relative Finder" database. In addition, there are many 3rd party tools designed to work with 23andme data (

    (Technically doing 23andme is not exactly the same as getting your whole genome sequenced; 23andMe "only" genotypes about a million markers. But again this will provide you most of the information that's currently useful. For actual complete sequencing, unless you have pressing medical need or other very specific interest, I'd probably suggest waiting a couple years for the price to go to down and the accuracy to improve.)