This is the blog for GW students taking Human Evolutionary Genetics. This site is for posting interesting tidbits on: the patterns and processes of human genetic variation;human origins and migration; molecular adaptations to environment, lifestyle and disease; ancient and forensic DNA analyses; and genealogical reconstructions.

GWHEG figure

GWHEG figure

Monday, November 30, 2015

Schmutzi - Software for Identifying Contamination in Ancient DNA

Earlier this month, Gabriel Renaud (Max Planck Institute, Leipzig) et al announced their release of software for estimating contamination in ancient human DNA. Named "Schmutzi," the program is written in C++ and is available for free download from the MPI Evolutionary Anthropology website. Analysing deamination---which is expected to be higher in ancient DNA than in modern---and the distribution of fragment lengths, Schmutzi compares hypothetical ancient mitochondrial genome reconstructions to possible contaminant modern genomes. The result is an estimation of contamination, as well as a consensus reconstruction of the ancient mtDNA genome. The program recognises differences between the sample aDNA and databased possible contaminant sequences one nucleotide at a time, with purported greater accuracy than any equivalent code to date; running on a PC, one to three hours are required for the analysis of one million base pairs.

Fig. 1 - A human (left) and Neanderthal (right) skull; the new program 'Schmutzi' from the Max Planck Institute may assist future studies of DNA from these close hominin cousins. (photo WikimediaCommons/DrMikeBaxter)

Source article published in Genome Biology Nov 2015

The [SNP's] That Shake the Barley

Scientists at the University of Minnesota and UC Irvine recently published their analysis of 803 regional domesticated varieties (landraces) and 277 wild strains of barley. The study finds that barley---a self-pollinating diploid cereal grain with a long history of human cultivation, ideal for genetic studies---cannot be traced to a single origin of domestication. Comparing single-nucleotide polymorphisms, Poets et al confirm that barley has been domesticated multiple times in human history, but over the course of extensive trade in this crop, regional cultivars have admixed with sympatric wild strains. Recent introgression is ruled out, so the ultimate conclusion falls to ancient gene flow. Each domestic cultivar exhibits admixture from multiple wild types, but with the greatest contributions being from geographically proximate wild barleys; domestic barley in the Levant shares more SNP's with Levantine wild barley, domestic barley in Central Asia shares more SNP's with Central Asian wild barley, and so on. Most fascinatingly, certain regions of the genome seem more prone to fixing this gene flow, raising the possibility that alleles gained from local wild barley are somehow adaptive. As the world's barley-growing regions are quite diverse, this makes sense and gives some insight to the nature of early human agriculture.

Fig. 1 - Negative correlation between admixture and geographic distance between cultivar and wild strain

Source article published in Genome Biology

Sunday, November 29, 2015

Where things get a little hairy: resurrecting the mammoth

In a review published earlier this month, UC Santa Cruz professor Beth Shapiro summarises present understanding of the possibility for "resurrecting" extinct species---that is, "de-extinction," and how and why it could become reality. Shapiro describes this process not so much about bringing back extinct species, but rather about re-introducing extinct traits via genome editing. As extinct taxa are recovered with ancient DNA available for analysis, the prospect of editing a closely related extant genome to possess autapomorphic sequences of the extinct species becomes possible; George Church's lab at Harvard University has already done this by replacing 14 loci in elephant cell lines with their mammoth-specific counterparts. If this could be done genome-wide, and a viable trans-nuclear embryo established in a host mother, a mammoth could hypothetically be "cloned" in much the same way as Dolly the Sheep.

Fig. 1 - Woolly mammoths as portrayed by the acclaimed Charles R. Knight

Even this would fall short of properly "resurrecting" the mammoth, however, because this synthetic creation would fundamentally develop and live in a different environment, in different social groups, with different stimuli and an ecological niche distinct from its prehistoric forebears'; changes genetic and epigenetic resultant from inconsistencies with the extinct mammoth's Pleistocene habitat would give rise to an animal likely different from the original (and unknowably so). Shapiro notes all this in accessible and fascinating detail, and describes de-extinction as more likely the use of genome editing to resurrect specific phenotypes---instead of a mammoth, a cold-adapted hairy elephant; instead of an aurochs, a large lean cattle with fewer maladaptive vestiges of artificial selection; instead of a Neanderthal, human cell lines with Neanderthal-derived genes relevant to health or disease. The entire prospect is fascinating and blurs the line between science-fiction and science (the animated sequence from the first Jurassic Park comes to mind), and raises riveting prospects and points of debate for the generations of scientists to come.

Fig. 2 - Remember me?

Source article published in Genome Biology Nov 2015

Ancient Infection - retroviral evidence preserved in modern primates

Scientists at the University of Michigan report the discovery of a novel retoviral derived sequences around the centromeres of extant catarrhines including humans. Named "K222," it exists as a single copy in baboons, gorilla, orangutans, and chimpanzees; but in humans, nine copies are found in the pericentromeres of nine different chromosomes. The introduction of this sequence into modern primate genomes can be traced to an ancestral retroviral infection of the germ line that occurred approximately 25 million years ago, after the split of Old World and New World primates---explaining the absence of this catarrhine symplesiomorphy among platyrrhines. The fact that only humans present multiple copies of K222, and all pericentromeric, suggests chromosomal recombination near the centromere within the last 6 million years of hominin evolution. As centromeres are generally considered immune to recombination during gametogenesis, pericentromeric recombination as a human autapomorphy could have interesting implications for studies in health and medicine.

Fig. 1 - Two catarrhines (photo credit Barry Bland)

Source from the journal Genome Biology

Friday, November 27, 2015

Population movements during the Neolithic revolution

An article published in the journal Nature this week reconstitutes the population movements in Eurasia at the time of the Neolithic revolution. The Harvard geneticist David Reich and his team sequenced the genomes of 230 fossils from that period.

The results suggest that lactase tolerance was strongly selected for as well as SLC22A4 which facilitates digestion of wheat and other crops. However, this gene is also linked to increased digestive disorders. In addition, skin color changes were documented. Light skin color would have been selected for when people started having a less vitamin D-rich diet with decreased meat consumption.  Lighter skin absorbs vitamin D from the sun better than darker skin. The study also reports data on changes in stature and population movements.

Thursday, November 19, 2015

Genome sequencing of primitive worms

This week, Nature published an article that details the genome sequencing of two genera of the marine invertebrate enteropneust, also called acorn worms. The research group that studied this animal were interested in their capacity to prevent water from entering their digestive tract while they breathe, through features called pharyngeal gill silts. The Japanese team behind the research found four highly conserved transcription factor genes that seem linked to pharyngeal gill silts. This study is important because it is rare that the genome of non-vertebrates and non-arthropods (insects, arachnids  etc) are sequenced, and thus this research provides depth in the understanding of the evolution of chordates. 

Acorn worms vary form a few millimeters to a few meters!

Recalibrating the Molecular Clock - New Approach for Calculating Mutation Rate

Most accurately approximating the human mutation rate is critical for estimating divergence time from the Panin lineage, and is necessary for understanding modern human genetic variation and the biology of genetic diseases.

A new publication in PLoS Genetics recalculates the human mutation rate by using the human recombination rate, thus avoiding reliance on an external calibration reference (eg fossil dates). They estimate a rate of 1.61 ± 0.13 × 10−8 mutations per base per generation (or 0.55 ± 0.05 × 10−9 per base per year, if generation time is 29 years) which suggests a Hominin-Panin population split time of 6.8 ± 0.6 Ma.

Wednesday, November 18, 2015

Antibiotics can alter your microbiota

The effect of culture on the microbiome (micro organisms commensal with humans) has been relatively well studied in deep time, most notably through the effects of diet change following the origin of farming. However, accelerated cultural changes in the last hundred years would have contributed to altering the microbiome in complex and poorly understood ways. A recent review article examines the impact of the use of antibiotics on human microbiota and suggests they might be one of the main drivers of microbiota diversity. According to this article, the use of antibiotics affects the composition and diversity of the microbiome during and after a treatment. An "unhealthy" microbiota is linked to a number of problems and diseases including allergies, depression and  inflammatory conditions.
Selective forces influencing the human microbiome through time

Tuesday, November 17, 2015

Genes and stature

Previous studies have shown stature is highly correlated with local environment and often has a genetic basis. A recent study of a hunter-gatherer pygmy population in Uganda has reported that the pygmy phenotype, small human body size, is associated with genomic regions that have previously been associated with stature variations. SNP comparisons of this population with other west African populations reinforced that the pygmy phenotype shows convergent evolution within Africa and is a result of a positive selection to better adapt to the demands of the environment.

The Geography of Genes

One can imagine that past cultures occupied geographically discrete areas, meaning that, genetically, you would be more likely to be more closely related to the people within your cultural group than those outside it. However, with the advent of modern transportation, gene flow between individuals of spatially separated groups is more likely than ever before. Given this, would people from various countries cluster into genetically similar groups? A study by Novembre and colleagues used a PCA on a sample of individuals genotyped at 500,568 loci, and revealed that not only did people of different European nationalities cluster into descrete groups, but the spatial structure of the clustering also resembles the political boundaries of Europe. If the ancestry of modern human populations (at least in Europe), reflects national boundaries, would it be possible to use aDNA to reconstruct the spatial boundaries of past cultures?

New data on the Denisovans 

Our knowledge of the Denisovans is mainly derived from genetic analysis of finger phalanx. A very recent publication on PNAS reported nuclear and mtDNA sequence data from two additional molars specimens (Denisova 4 and Denisova 8) assigned to the Denisovans. The two molar specimens are reported to be older than 50,000 years based on radiocarbon dating and are morphologically distinct from Neanderthals and modern humans. The study confirmed the low genetic diversity of Denisovans proposed in earlier studies, although higher compared to Neanderthals,  and indicated a longer settlement, and possibly a wider geographic coverage of the Denisovans in the region inferred from the low substitution in the Denisova 8.


Ancient DNA analysis reveals the nature of early cattle domestication

Human domestication of the prehistoric wild aurochs, Bos primigenius (pictured below), is considered one of the most important milestones in the rise of modern human culture. Dated to approximately 10,000 years before present, the first domesticated cattle represent humans' early foray into the realm of keeping livestock over hunting and gathering--in time, the stability afforded by this and other agricultural practices would allow for the establishment of cities and towns.

A new analysis of ancient DNA from a 6,750-year-old British wild aurochs offers insight to the early days of domestic cattle's divergence from the aurochs. The two living species of domestic bovines--cattle (Bos taurus) and zebu (Bos indicus)--were domesticated respectively in the Middle East and South Asia from populations of Bos primigenius, but did early cattle and contemporaneous wild aurochs continue to interbreed? The ancient genome analysis suggests that by and large the answer is no, as most living breeds of cattle show little evidence of mtDNA similarity with the aurochs and nuclear DNA likewise suggests minimal admixture. However, the aurochs genome shows similarity and significant evidence of gene flow with traditional Irish and British cattle breeds such as the Highland, Kerry, Dexter, Welsh Black, and White Park. So while sampled modern African, Asian, Middle Eastern, and mainland European cattle show minimal evidence of inward gene flow from prehistoric aurochs, breeds endemic to the British Isles stand anomalous. The researchers suggest this may be evidence of early British herders purposefully restocking with indigenous wild aurochs, or at least that cattle-aurochs hybrids survived and produced lineages contributing to modern-day populations. The study provides a fascinating look at the earliest days of one of humanity's most pervasive modern behaviors and widespread cultural phenomena, and paves the way for future analyses of anthropogenic artificial selection colliding with the natural world.

Study published in Genome Biology (2015).

Friday, November 13, 2015

Could Neolithic 'cultural fitness' have caused a male reproductive bottleneck?

Earlier this year, a paper published in Genome Research sequenced a sample of globally diverse Y chromosome data and found a bottleneck within modern Y chromosome lineages that occurred about 8 - 4 ka. The results of this study suggest that at this time the effective population size of females was 17 times higher than that of males.

Since the timing of the bottleneck coincides with periods of rapid cultural and technological change, including numerous agricultural innovations, the researchers claim that the reduction in male reproductive diversity could be a signal that fewer males were reproducing. A potential explanation for this would be that the men were competing culturally for women, rather than mate choice being random or driven by biological fitness.

If this hypothesis holds up after further genetic and archaeological research, this would be a good example of cultural practices driving large scale genetic changes in ancient human populations.

Wednesday, November 11, 2015

Gee thanks, Dad!

A study published three days ago in the journal Science investigates the hypothesis that physiological effects of life experiences may be inherited by a man's offspring, through epigenetic markers maintain in his sperm. As CpG methylation of DNA has shown to be seldom passed from germ-line cells to sperm and subsequent posterity, this team of scientists from Canadian and Swiss universities instead looked at methylation of the histones that DNA intertwines itself with in vivo. In human sperm, when germ-line cells undergo the final stages of spermatogenesis, 85% of histones are removed from the nuclear DNA so that the genetic material may condense itself to fit in the tiny sperm cell; in the remaining 15% of histones, however, methylation or demethylation of the histones may have effects on offspring and possibly even subsequent generations. Here, the researchers overexpress human KDM1A histone lysine 2 demethylase in male transgenic mice and find that offspring overwhelming present birth defects, mortality, and changes in the their transcriptome; even if an offspring was born from a sperm that happened not to carry the transgenic KDM1A, the mouse exhibited changes in histone methylation resultant from KDM1A overexpression in the parent testis. This may bear implications for humans as well---our fathers' and even our grandfathers' mutagenic experiences, if affecting epigenetic controls in reproductively relevant tissue, may have ramifications in our own bodies today.

Fig. 1 - "The Force is strong in my family." - Luke Skywalker commenting on the heritability of communing with the Force, after learning that Darth Vader / Anakin is his father. This lends to the hypothesis that strength in the Force is related to epigenetic events in the paternal lineage.

Effects of the epigenome on the pathogenesis of schizophrenia

Schizophrenia is a debilitating psychiatric disorder, in which the affected patient experiences hampered ability to perceive reality. It is among the most researched and least understood mental diagnoses, particularly because its pathogenesis lies at the intersection of polygenic heritable traits and diverse environmental stimuli. Kimberly Shorter and Brooke Miller, from the McKnight Brain Institute and Depts. of Psychiatry and Medicine at the University of Florida, recently reviewed current understanding of epigenetic influence on the schizophrenic phenotype. As it turns out, the neurochemical situation behind schizophrenia is still largely unknown, with evidence for the role of the neurotransmitter dopamine based on the efficacy of antipsychotic drugs that target it; however, genome-wide association studies show little difference between dopamine-related genes in schizophrenic (SCZ) and non-SCZ individuals. The search for differences must then turn to the epigenome: the collective regulation and de-regulation of select regions of DNA, responsible for the same species-wide genome presenting itself differently in 7 billion people worldwide.

"Epigenetic modifications" may entail methylation and demethylation of strand of DNA, respectively preventing or allowing transcription of genes into RNA/protein; histone modifications, such as acetylation; non-coding RNA; and coiling/super-coiling of chromatin, i.e. how tightly packed the cell stores various stretches of its DNA. Genes related to schizophrenia include brain-derived neurotrophic factor, glucocorticoid receptor, glutamate decarboxylase 1, and reelin---in SCZ patients, all four of these genes are hypermethylated and heavily down-regulated. Interestingly enough, some of these genes have also shown association with substance dependency, autism, bipolar disorder, and Alzheimer's disease. So while the connection to neural function is clear, this case highlights the frustrating truth that many symptomatically different brain disorders seem to stem from overlapping causes. Another gene, ST6GALNAC1, associated with stress-activated kinase signaling, is severely de-methylated in SCZ post-mortem brain tissue, resulting in overexpression of the gene in patients.

Histone acetylation, an epigenetic mechanism that causes DNA/chromatin to un-coil and loosen in structure, allowing transcription, is also implicated by the pharmacology of schizophrenia treatment. A common treatment for both SCZ and bipolar disorder (BPD) is the drug valproic acid (VPA); VPA functions to inhibit the enzymes that de-acetylate histones, thereby maintaining acetylation and the expression of genes. One histone deacetylase in particular, HDAC1, is overexpressed in SCZ patients and leads to overexpression of genes particularly located around histone 3. For this reason, Shorter and Miller propose investigation of more specific inhibitors that target only the HDAC's associated with schizophrenia; this way, the side effects of "off-targeting" by drugs such as valproic acid may be reduced.

The review is highly informative and provides exciting insights on the use of epigenome studies to better human health---in time, our understanding of epigenetic diversity and phenomena may be able to lessen the suffering and improve the lives of many who struggle with mental health.

Fig. 1 - Self-portrait by Craig Finn, a schizophrenic man (PLoS Med, 2005)

Is it really the environment that causes alcoholism?

For a long time, it has been hard to determine if there is an exact gene prone to causing alcoholism. It is in fact a myth that one single gene that leads to addiction or 'addictive personalities.' There is no single gene or factor that causes alcoholism or any type of addiction to substances but in fact a variety of factors from influences in early childhood to a number genes or lack of genes that can lead to addictive tendencies. Genetics clearly play a huge factor in addiction. Children who's biological parents are prone to alcoholism are 3-5 times more likely to develop an addiction even when raised by parents with non-alcoholic traits. One gene attributed to alcohol prevention is ALDH2, which has statistically demonstrated that the individual carrier is nine times less likely to develop alcoholism. Still, environment shows a clear impact on the development on addictive behaviours. For example, in Japan between 1972-1996 alcoholism has risen from 2.5%-13% in the overall population as drinking has become more as heavy-drinking culture has developed among business men. Epigenetic factors are also huge in the development of alcoholism as they can switch on regulatory factors of the brain that can lead to problems later in life as the child develops without the genes switching off when they should. Childhood trauma such as severe stress early in life can double the risks of developing alcoholism.

Tuesday, November 10, 2015

Archaic Epigenomes

The ongoing research on the ancient DNA of hominin fossil is continuing to provide insight into the biology of our extinct ancestors. A study by Gokman et al. (2014) details the reconstruction of DNA methylation maps of both the Neanderthals and Denisovans, and provides insight into the epigenetic basis for phenotypic differences between present-day and archaic humans. However, just like nuclear aDNA, traditional methods designed to measure full methylomes are not applicable to archaic samples due to the postmortem degradation of the DNA. Despite this, Gokman and colleagues note that unmethylated cytosines decay to uracils, whereas those that are methylated decay to thymines. Cytosines that were methylated prior to death should exhibit a higher fraction of reads with thymine in comparision to unmethylated cytosines. Therefore, the ratio of C to T substitutions provides a robust proxy for methylation level in aDNA. Using this proxy, the study revealed around 1100 differentially methylated regions that were then assigned to a species. Among these differentially methylated regions was a HOXD cluster, a key regulator in limb development. In Neanderthals and Denisovans, HOXD9 and HOXD10 were hypermethylated, whereas they are hypomethylated in present-day humans. These differentially methylated regions imply regulation of HOXD9 and HOXD10 consistent with morphological differences between Neanderthals and modern humans.

Hormones and social cognition

Oxytocene is a hormone that is believed to play a big role in human behaviors. DNA Methylation, on the other hand is one of the mechanisms that affect gene expression. A recent study on PNAS used brain imaging of 98 individuals to study the effects of Methylation of OXTR, which is a gene that codes for the receptor protein of oxytocene molecul, to better understand the effects of oxytocin. The study has shown that methylation of the OXTR affects areas that are important in face perception and emotional processing and that it is associated with increased reaction to the stimuli such as fear and anger. Results also showed decreased functional connectivity in different brain regions supporting earlier assumptions that oxytocene is associated with social cognition.

Monday, November 9, 2015

"A Very Muscular Baby Offers Hope Against Diseases"

Occurrence MSTN polymorphisms across human populations.

To follow up my "Super Muscular Puppies" blog post, this articles concern natural variation of the myostatin gene (MSTN) in human populations. Mice with knocked out MSTN show massive increases in muscle cell number and size. Seven years after the publication of this finding, a case study of a child with a homozygous loss of function mutation in MSTN was documented (also covered in an article with the fantastic title, "A Very Muscular Baby Offers Hope Against Diseases"). This child, a German boy named Liam Hoekstra, showed a similar phenotype to the KO mice in that he had visibly increased muscle mass relative to a typical newborn. By the age of 3 years, it was claimed that Liam had unusual strength for a child his age (though this was never formally tested). Studying cases like that of Liam holds promise for developing muscular wasting disease therapies.

In 2006, another study tested for evidence of positive selection of the MSTN gene across human populations and between species. A positive selection analysis calculated dN/dS ratios (the proportion of non-synonymous mutations to synonymous mutations) across 23 vertebrates (mostly mammals) and found a high degree of conservation of this gene. However within human populations, several allelic variants produced by SNPs were present in a high proportion (up to 31%) in sub-Saharan populations. Rather than resulting in a complete loss of function, these mutations were associated with a partial reduction in the ability of MSTN to inhibit muscle growth. However the exact functional consequences of these mutations in these individuals is still unclear, as a previous study failed to find associations between MSTN polymorphisms and muscular strength. The authors of the 2006 article claim that the mutations seen in sub-Saharan populations may result in increases muscle mass (not tested in these individuals) which could protect against famine (a counterintuitive explanation given the high metabolic cost of muscle tissue).

Super Muscular Puppies

Myostatin is a protein that acts as a negative regulator of muscle cell (myocyte) production and growth. However in rare cases, inactivating mutations to the myostatin  gene (MSTN/GDF8) can increase in the number (hyperplasy) and size (hypertrophy) of myocytes, resulting in a 200-300% increase in muscle mass relative to controls. This has been demonstrated experimentally by knocking out MSTN in "mighty mice".

Similar mutations to MSTN have also been demonstrated to be the mechanism underlying massive muscle increase in domesticated animals selectively bred to be "double-muscled", such as Belgian Blue cattle and whippets dogs.

Recently, a group of researchers were able to create transgenic beagles using CRISPR to selectively mutate the MSTN gene. This resulted in two puppies with different sets of MSTN mutations. Puppy #11 (dubbed Tiangou, after the Chinese mythical heaven dog) demonstrated similar muscle mass to littermate controls, due to being chimeric (i.e. some cells carried the mutation while others did not). However puppy #5 (dubbed Hercules after...Hercules) was shown to have biallelic MSTN mutations in all tissues resulting in the double-muscled phenotype observed in the aforementioned domesticated breeds.

Authors of this recent publication note that this could serve as an alternative model in the development of therapies for muscle wasting disorders, but also mention the potential to create "new strains of dogs with favorable traits for other purposes." The implication of designer pets brings to mind ethical concerns. Though essentially, all pure-bred dogs are freakish inbred wolves, mutated through the generations of artificial selection instead of direct genetic manipulation