Research

The Keinan lab aims to improve the search for complex disease genes and genes underlying other complex traits, with the key driving hypothesis being that characterizing human population genomics can inform the design and analysis of medical genetic studies. This hypothesis received fresh support from the lab's work on the effect that recent human explosive population growth has had on the accumulation of rare genetic variants and the extensive implications of that discovery for gene-disease association studies (Keinan & Clark, Science 2012). Hence, the lab studies how demographic history and natural selection have shaped patterns of human genetic variation, and translates that knowledge to the study of the genetic basis of complex human diseases. Members of the lab come from varied backgrounds, including in computer science, statistics, genetics, genomics, physics, anthropology, and biology, which enables the collaborative development of computational and statistical methods, their efficient application to large-scale, genomic data sets, and the interpretation of discoveries in light of gene function and anthropological evidence.

Recent Areas


The X-Factor of Complex Disease

Methods, software, and extensive application for studying the X chromosome in association studies

 

  

 
The X chromosome plays an important role in human disease, especially those with sexually dimorphic characteristics. Analysis of X requires special attention due to its unique inheritance pattern leading to analytical complications that have resulted in the majority of GWAS either not considering or mishandling it with tools designed for non-sex chromosomes. We overcame many of the analytical complications by developing an array of X-specific methods that span all stages of GWAS, from genotype calling, through imputation and extensive QC, and to statistical association testing. Specifically, we develop new types of association tests, some of which applicable uniquely to the X chromosome. We implement the analysis pipeline and all methods as part of a publicly available software, XWAS (chromosome X-Wide Analysis toolSet). We apply these to conduct X-wide association studies in dozens of GWAS, with focus on autoimmune diseases, risk factors of coronary artery disease, and psychiatric disorders, all of which are very different between males and females. We discovered and replicated many novel significant X-linked associations, e.g. (i) variants in CENPI as contributing, with different effect sizes in males and females, to the risk of three different autoimmune diseases. Other, autosomal genes in the same family as CENPI have previously been associated to other autoimmune diseases; (ii) ARHGEF6 to Crohn's disease, and replicated in ulcerative colitis, another inflammatory bowel disorder. ARHGEF6 has been shown to interact with a gastric bacterium that has been associated to IBD. (iii) Significantly increased variance of systolic blood pressure in females that are heterozygous for a variant that might regulate ATRX, a gene that has been previously associated with alpha-thalassemia. With the availablility of the XWAS software package, we hope to bring the X chromosome to the GWAS era by enabling other researchers to include it in their studies.
 
 
Some related publications:

Nutrigenomics – Genetic adaptation to changing diet during human evolution and its impact on complex disease

We are interested in identifying genes and genetic variants that have helped human populations to adapt to historical types of subsistence (diets) For instance, diets can be considered by the relative consumption of plants (fruits, vegetables, seeds), meat, and different types of seafood. As different diets were prevalent in different times and places, they led to individuals today having genetic differences that lead to differences in nutrient metabolism, metabolic disease risks, and risk of many other diseases. Our nutrigenomic research involves studying adaptive variants by using and developing population genetic methodologies, based on both ancient DNA from different epochs and DNA from extant populations worldwide. Beyond testing for adaptation of genetic variation, we conduct quantitative genetics analyses to reveal the phenotypic relevance of genetic variation using cohorts of many individuals and interventional studies that test individuals before and after a change in their diet. Functional genetic evaluation of molecular mechanisms, along with nutritional and anthropological interpretations, allow us to verify our results. These also allow positioning our results towards dietary recommendations based on an individual’s genetics, diet, the population of origin, and the interaction of these.

 

Thus far, our unique discoveries involved the fatty acid desaturases (FADS) genes, whose enzymatic products catalyze the biosynthesis of the active forms of omega-3 and omega-6 (long-chain polyunsaturated fatty acids such as ARA, EPA, and DHA) from inactive (short-chain) forms as those consumed from pants. We found that alternative alleles of the same genetic variants in FADS genes have both been adaptive; One allele in farming populations with predominantly plant-based diets and the other allele in hunter-gatherer populations with predominantly animal/seafood-based diets. The former increases the expression of FADS1, thereby making the biosynthesis more effect and the latter decreases the expression. This discovery and additional ones we made concerning FADS have been widely (and often wildly) reported by the media. A few sound written accounts can be found in the Washington Post and Le Monde, a nice TV coverage on CBS News, and a podcast with humanOS Radio. The Cornell Chronicle also wrote great pieces of two of our papers (Eating green could be in your genes and Modern European genes may favor vegetarianism). Our most recent results serve as the basis of a direct-to-consumer genetic testing product by Insitome, based on the Helix DNA kit. For more, check out these past news posts on our website: Our paper on selection of a "vegetarian allele" piqued the interest of many in the media and Our recent publication in Nature Ecology and Evolution calls for Personalized Nutrition.

 

  

 

In ongoing and planned research projects, we continue to refine the picture of the evolution of the FADS genes worldwide, their effect on health, and making small dietary recommendations based on our results for improving one’s well-being.

This research points to the key role that diet has had in recent human adaptation. Hence, we are also developing and applying new methods for detecting additional genes with signals of natural selection by integrating ancient and modern DNA.

 
Some related publications:

Recent human population growth

  

 
Human populations have experienced explosive growth since the Neolithic revolution. The Keinan lab studies the effect that this unique demographic scenario has had on patterns of genetic variation, as well as use genetic data to gain new insights into the way different human populations have grown and spread. They characterize how recent growth increases the abundance of rare genetic variants and how it affects the workings of natural selection. Based on these population genetics insights, they study how recent growth has shaped the genetic architecture of complex disease and, as a consequence, what methods for gene-disease association testing would be most powerful for associating rare variants with disease risk.
 
 
Some related publications:

Contrasting patterns of genetic variation between chromosome X and autosomes

 

The genetic diversity of chromosome X is expected, under equilibrium conditions, to be three-quarters of that of the autosomes in a population with equal numbers of males and females. However, deviations from this ratio can result from at least four factors known to have been prevalent in human ­history: (i) sex-biased demographic events leading to different ­ effective population sizes of males and females; (ii) changes in population size over time; (iii) natural selection, which also affects chromosome X differently; and (iv) ­differences in mutation rates between sexes or between chromosome X and the autosomes.  We are interested in understanding how these factors have shaped the varying patterns of variation of X and A in different human populations.

 

Some related publications:


The genetic basis of complex human disease and other complex traits

 

 

Genome-wide association studies (GWAS) have provided important insights into the genetic basis of complex human diseases and traits.  At the same time, the current generation of GWAS has left us with the challenge of "missing heritability," whereby for most complex diseases and traits only a relatively small fraction of estimated heritability has been explained to day.  We develop and apply statistical and computational methods for detecting the contribution of gene-gene interactions (epistasis), rare genetic variants, and the sex chromosomes to complex disease risk, thereby elucidating the role they play in explaining missing heritability.  This work is in collaboration with several GWAS consortia.  Further, we make the software that implements our methods publicly available, which allows application in additional studies, thereby accelerating the understanding of complex disease etiology.  Our main foci are autoimmune diseases and lipid levels as risk factors of coronary artery disease.

 

Some related publications: