Congratulations to the winners of the Editors’ Choice Awards for outstanding articles published in GENETICS in 2021! The journal’s Editorial Board considered a diverse range of articles, finding many papers worthy of recognition. After much deliberation, they settled on one exceptional article for each of the three award categories: molecular genetics, population and evolutionary genetics, and quantitative genetics. Check out some of the best GENETICS had to offer in 2021, and be sure to browse the full Spotlight collection.

EDITORS’ CHOICE AWARD IN MOLECULAR GENETICS

Neurogenesis in the adult Drosophila brain

Kassi L Crocker, Khailee Marischuk, Stacey A Rimkus, Hong Zhou, Jerry C P Yin, Grace Boekhoff-Falk

GENETICS Oct 2021, 219(2), iyab092, https://doi.org/10.1093/genetics/iyab092

Crocker et al. describe the Drosophila central brain as a new model in which to investigate adult neurogenesis. The authors observe a significant increase in the number of proliferating cells following injury; they detect new glia, new neurons, and the formation of new axon tracts that target appropriate brain regions. The authors anticipate that this paradigm will facilitate the dissection of the mechanisms of neural regeneration and that these processes will be relevant to human brain repair.

EDITORS’ CHOICE AWARD IN POPULATION AND EVOLUTIONARY GENETICS

The timing of human adaptation from Neanderthal introgression

Sivan Yair, Kristin M Lee, Graham Coop

GENETICS May 2021, 218(1), iyab052, https://doi.org/10.1093/genetics/iyab052

Some Neanderthal-introgressed alleles in modern human populations were adaptive; however, the context in which they provided a fitness advantage is unknown. Yair, Lee, and Coop develop a population genetic method that uses ancient DNA and the hitchhiking effect to determine when natural selection favored the spread of Neanderthal-introgressed alleles. They identify regions of the genome in which Neanderthal alleles were immediately adaptive and others in which there was a significant time lag between admixture and the allele’s rise in frequency.

EDITORS’ CHOICE AWARD IN QUANTITATIVE GENETICS

Why genetic selection to reduce the prevalence of infectious diseases is way more promising than currently believed

Andries D Hulst, Mart C M de Jong, Piter Bijma

GENETICS April 2021, 217(4), iyab024, https://doi.org/10.1093/genetics/iyab024

Quantitative genetic analyses of binary disease status indicate low heritability for most infectious diseases, suggesting that the potential response to selection in disease prevalence is limited. By integration of quantitative genetics with epidemiological models, Hulst, de Jong, and Bijma show that the typical low heritability values of disease status correspond to a substantial genetic variation in disease susceptibility and to a large potential response to selection. Positive feedback mechanisms occurring in disease transmission are crucial for this response and even make eradication of infectious diseases possible. However, current quantitative genetic models ignore these feedback effects and thereby underestimate response to selection in disease prevalence.

Guest post by C Brandon Ogbunu.

2022 marks the return of the Population, Evolutionary, and Quantitative Genetics (PEQG) Conference, organized by the Genetics Society of America. Part of the meeting’s popularity stems from being one of the few conferences that brings together leading thinkers in subfields of genetics that don’t typically overlap, across a range of model organisms, united by methods and perspectives.

The meeting, which will take place June 7-10 in Pacific Grove, CA, at Asilomar Conference Grounds, is well-known for its structure: a combination of keynote addresses, awards, and short talks of various kinds. One of the key aspects of this structure is the session chairs: junior scientists who have established themselves as leaders in the various areas of population, evolutionary, and quantitative genetics. During the meeting, they each chair a session full of talks, and give 30-minute talks of their own during the final keynote session. The session chairs provide an opportunity for us to view the present and future of the field.

The 2022 session chairs promise to deliver on this tradition, featuring a tremendous lineup of thinkers who study problems as diverse as speciation genomics in plants to epistasis in human genomic data sets. This specific collection of speakers displays both breadth and depth, and so the chair keynote session promises to excite.

Below I will highlight these session chairs, commenting briefly on why I am personally so excited to hear about their work.

Nancy Chen

Evolutionary biology is, in part, a science that is defined by information from the past, but how do we use it to ask questions about contemporary evolution in natural populations? These are the questions of the “Pop Gen Chen Lab,” run by Nancy Chen. The lab addresses questions and utilizes tools to think about contemporary questions in short-term evolution, and how genetic variation is maintained in contemporary populations. In addition, the Chen lab makes use of the Florida Scrub Jay (a very compelling and well-studied system) to study population decline. The Chen lab has also generated an extremely useful list of resources on issues related to diversity, equity, and inclusion, and continues to be a leading voice on these matters in the population genetics community. 

Lorin Crawford

Though he was raised in southern California, Lorin Crawford will come to PEQG from balmy New England, where his research program is sprawled out between Microsoft Research in Boston and Brown University (in Providence) where he is the RGSS Assistant Professor. It is difficult to fully capture the richness of his research program. He utilizes advanced statistical and machine learning approaches to directly address provocative questions in population genetics. For example, he has pioneered statistical tests that can be used to detect pairwise epistasis between mutations in large genomic data sets. In addition, his work dissects the architecture of complex traits. Lastly, Crawford has recently begun to explore the ethics of genomics evolution. Recent work in this realm has challenged notions that are used to characterize populations, such as “transethnic.”

Rafael Guerrero

From North Carolina comes Rafael Guerrero, an Assistant Professor at North Carolina State University. He runs a program that develops tools that have already transformed our approach to classical questions in population and evolutionary genetics and explores the questions directly relevant to practical problems in biomedicine and bioengineering. In the former sense, Guerrero has done groundbreaking work on chromosome evolution and hybrid incompatibilities in light of speciation genetics, both central and critical questions in evolutionary genetics. In the latter sense, Guerrero’s mastery of theoretical tools has allowed him to explore areas such as the genomics of adverse pregnancy outcomes, and the physiological determinants of epistatic interactions as they manifest in the evolution of antibiotic resistance.

Priya Moorjani

The Genetics Society of America is well-known for its commitment to model systems research and has long championed its importance. But it also recognizes the importance of human genetics and evolutionary biology, not only because we…are humans, but also because human evolution is an amazing problem space for cutting-edge questions in evolutionary and population genetics. Few scientists are doing more exciting work in this area than Priya Moorjani. Moorjani uses statistical and computational approaches to understand the role of genetic variation in human evolution, demography, and mapping disease risk alleles at the University of California, Berkeley. Moorjani has also investigated fundamental questions in primate evolution, such as the proper estimation of mutation rates. Moorjani has mastered the art of transforming a species that we all care about–Homo sapiens–into a model system in evolutionary genetics. 

Rori Rohlfs

As Assistant Professor at San Francisco State University, Rori Rohlfs won’t need to travel especially far to get to Asilomar, but everything with Rori is an intellectual expedition. Rohlfs runs an exciting program that has examined everything from the evolution of gene regulation to critical statistical questions relevant to genomic testing and forensics. Rohlfs has accomplished this while also being a widely recognized teacher and mentor. Lastly, Rohlfs was one of the corresponding authors on an outstanding 2019 study published in GENETICS that analyzed early population genetics literature and identified the many women that were often denied proper credit for their participation.

Daniel Runcie

When I teach evolution, I often discuss a 2018 study that estimated the biomass of living things on earth, organized by different taxa. Though I do not study plants, I often use it to explain that when it comes to life on earth, plant life is the heavyweight champion. Daniel Runcie runs a thrilling research program that attempts to understand how and why plants are so successful, and especially questions related to genetic variation and phenotypic plasticity. The Runcie lab attempts to identify pathways and networks related to how plants respond to a dynamic environment. One of the reasons that plants have been so successful is their ability to respond to change. The Runcie lab uses a host of tools—statistical, network, and ecophysiological—to understand these questions.

Learn more about the #PEQG22 Session Chairs, as well as Invited Speakers, on the conference website. Registration is open now.

About the author

C. Brandon Ogbunu is Assistant Professor in the Department of Ecology and Evolutionary Biology at Yale University and one of the organizers of the 2022 Population, Evolutionary, and Quantitative Genetics Conference.

Researchers used a forward genetic approach to identify genes that affect a social behavior in honey bees.

For more than 30 years, honey bee geneticist Robert E. Page, Jr. and his colleagues have sought the genes that influence a colony trait that only emerges from interactions between thousands of individual bees — a social phenotype.

Such traits are notoriously difficult to study. As hard as it is to disentangle a gene’s effect on behavior from environmental influences, these challenges are greatly multiplied by the complex interactions between genetically different individuals forming and altering their own social environments. As a result, mapping quantitative trait loci (QTLs) for social behavior long seemed out of reach. A recent paper published in GENETICS surveys the three decades of work that culminated in identifying genes that affect a complex, socially-regulated foraging behavior.

“Everything flowed from our observations of the behaviors that were associated with foraging behavior and division of labor and the resulting impact on the amount of pollen stored in the comb,” says Page, a researcher at Arizona State University. “Over time, we adapted new technologies to ask questions about those phenomena in different ways. By bringing together a host of different toolkits, we were able to build a story of the underlying genetic basis of a very complex social trait.”

Pollen hoarders

In a honey bee colony, some of the bees specialize in collecting pollen and hoarding this protein-rich food in wax cells near the “nursery” where eggs and larvae develop. Some of the bees eat the pollen and produce glandular secretions to feed the larvae; in turn, the larvae produce pheromones that stimulate foragers to collect pollen.

Despite the complexity of factors involved — thousands of individual bees, larvae, their interactions, and their environment — the total amount of pollen stored in the colony is a regulated trait. In a previous study, ASU reseacher Professor Jennifer established this by adding and removing pollen from honey bee colonies and observing the changes in foraging behavior. Each colony had a set level of stored pollen that the bees collectively sought to maintain; when researchers added pollen, foraging decreased until excess pollen had been consumed, and when researchers removed pollen, foraging increased until pollen again reached the colony’s standard level.

Using selective breeding, Page’s team generated strains of honey bees that substantially differ in amount of pollen stored within just three generations. “People have tried for decades to breed bees that store more honey, but honey storage is a very sloppily regulated trait,” says Page. “As long as nectar is available, bees will bring it back and stick it anywhere they can find space in the hive. In contrast, pollen storage offers excellent consistency of measurement.”

Phenotypic and genotypic analysis

Using a wide range of methods, Page and his colleagues studied the phenotypes and genotypes of these strains for 42 generations of selection. Phenotypic mapping revealed that bees in colonies that store more pollen are likely to have more ovarioles (the egg-producing structures in the insect ovary) and to be more sensitive to sugar than those from the colonies with less pollen. One research collaborator, Ying Wang, went so far as to painstakingly graft ovarioles from worker bees into recipients from a colony that produced less, which in turn affected the recipients’ behavior.

When they began to map genetic trait determination, the researchers expected that individual behavior would be more selectable than complex social traits playing out across large groups. Instead, they were surprised to find that genotype explained roughly 41 percent of colony variance in stored pollen but only two percent of individual variance in pollen collection.

Digging down to the genetic level, Page and his colleagues performed QTL mapping for the social phenotype of pollen hoarding along with individual foraging behavior, physiology, and anatomical traits. From the gene lists for each QTL they identified candidate genes of interest based on the phenotypic architecture and assessed them using expression assays and gene knockdown. Ultimately, they identified three genes of special interest, all of which have some association with ovary size. Future studies will involve examining the effects of these genes in developing larvae more closely.

A vertical approach

Page attributes his success in part to his tenacity in pursuing a “vertical approach”—interrogating a single question at every level of influence, starting with social interaction and working down to examine individual behavior, morphological differences, physiology, developmental processes, and genetic variation.

Pursuing this vertical approach over so many years involved the work of countless expert apiculturists, laboratory technicians, students, and fellow researchers alongside Page. “Our success came from combining the right people with the right tools as they became available,” he says. “Everyone who came through my lab made it better and brought something new.”

CITATION:
Societies to genes: can we get there from here?
Robert E Page, Jr.
GENETICS 2021; iyab104
https://doi.org/10.1093/genetics/iyab104

Researchers identified dozens of quantitative trait loci controlling important traits in Cannabis sativa.

In 2014, United States federal law changed to allow scientific research on Cannabis sativa in states with regulated hemp programs. This legal shift opened the door to research that had previously been slow and difficult due to regulatory hurdles and funding challenges.

A new study published in GENETICS capitalized on this new opportunity and identified 69 quantitative trait loci (QTLs) that are responsible for variation in key agronomic and biochemical traits in C. sativa. This research is a step towards understanding the genetic control of complex traits in hemp and will inform future investigations into the overall evolution and function of complex traits across multiple species.

Hemp is grown for a wide range of commercial uses, including in building materials, textiles, and composite plastics, food and drink, animal feed, and pharmaceutical cannabinoid products, says study leader John McKay of Colorado State University. “It’s important to understand the genes controlling this plant as a crop, and it’s also interesting from a fundamental evolutionary standpoint.”

Identifying QTLs in a non-model species

“Scientists have previously made progress in identifying the genetic basis of complex traits in model and crop species—our team wanted to ask those questions in C. sativa because it’s an understudied species,” says McKay.

Today’s genome sequencing and assembly technologies are largely species-agnostic. As a result, scientists can now take bioinformatic and statistical genetic tools that were initially developed and tested in fruit flies and Arabidopsis and increasingly apply them to non-model species.

McKay and his team developed an F₂ hemp population by crossing two phenotypically distinct varieties—a tall, late-flowering cultivar bred for fiber production and a shorter, early maturing hemp that was bred for both fiber and grain crops. They then used whole genome sequencing to map QTLs associated with traits of interest, such as grain yield and stem biomass, along with 17 biochemical traits. These included levels of cannabidiol and terpenes, which have medical uses, and THC, which is the psychoactive component of marijuana. THC is strictly regulated in the US, and THC levels in industrial hemp must be below 0.3% to remain legal crops.

Most QTLs they identified clustered into one of four genomic regions, suggesting that much of the difference between the two varieties is due to a small number of genes that have large pleiotropic effects. Two candidate genes emerged that may underlie some of these clusters: the homolog of an Arabidopsis transcription factor gene called TINY may be associated with a cluster of agronomic traits, and the gene for olivetol synthase appears to underlie variation in a cluster of biochemical traits, consistent with the enzyme’s role in cannabinoid synthesis. The researchers functionally validated the olivetol synthase candidate by expressing the two hemp alleles in yeast. They found that the allele from the low-cannabinoid cultivar produced less olivetol in the yeast expression system, supporting the hypothesis that allelic variation at this gene plays a role in the observed phenotypic variation.

The researchers also observed epistatic interactions between some of the QTL clusters, further complicating attempts to elucidate any one trait’s exact genetic underpinnings.

“This study definitely adds to the broader conversation about complex traits,” says McKay. “For example, everyone agrees epistasis exists, but breeders and geneticists like to argue about whether it’s important to include in prediction models. Documenting additional cases like this in which epistasis contributes to variation adds to our understanding of the basis of complex traits.”

Traits are complicated but still predictable

The results of this latest study contradict a paper from 2003 that concluded that variation in cannabinoid production is controlled by a single genetic locus. The team from Colorado State University identified at least four loci controlling variation in these chemotypes.

“The field of genetics has always been a friendly place to hypothesize that something—anything—is polygenic,” laughs McKay. “Finding multiple loci controlling a single biochemical trait wasn’t surprising to me, because the abundance of any molecule can be influenced not only by the pathway that makes that molecule but also the ones that influence the cells and machinery that contribute to the process.”

However, despite overturning assumptions of one-to-one genotype-phenotype interactions, McKay emphasizes that he still views the hemp traits in the study as predictable. Future grants would allow research groups like his to dive deeper into the adaptive value of cannabinoids in hemp plants and create more precise genetic manipulations of key traits of interest. Researchers are eager to see policy informed by scientific understanding of the factors that predictably affect cannabinoid content and other traits in hemp crops.

CITATION:
Quantitative Trait Loci Controlling Agronomic and Biochemical Traits in Cannabis sativa
Patrick Woods, Brian J. Campbell, Timothy J. Nicodemus, Edgar B. Cahoon, Jack L. Mullen, and John K. McKay
GENETICS 2021; iyab099
https://doi.org/10.1093/genetics/iyab099

The James F. Crow Early Career Researcher Award recognizes outstanding achievements by students and recent PhDs presenting their work at the Population, Evolutionary, and Quantitative Genetics (PEQG) Conference, which was part of TAGC Online in 2020. The 2020 winner and finalists for this prestigious PEQG award spoke in a high-profile session at the conference. Check out the recording below!

Winner

Carl Veller, Harvard University 

Finalists

Learn about some of the work that graduate students, postdocs, and early career faculty are contributing to the field of genomic prediction.

Since 2012, the GSA Journals have published a series of papers focused on genomic prediction. We’re excited to announce a newly-organized Series page that makes it easy to navigate the extensive collection of genomic prediction papers published at GENETICS and G3. We’d also like to introduce you to a few early career scientists currently working in the field and to give you a glimpse into the types of research they do.

Antoine Allier
Graduate student, INRA
Le Moulon, Alain Charcosset Lab

“My current research aims at optimizing the management of genetic diversity in breeding programs using genomic selection. In particular, I am working on the prediction of cross variance and genetic diversity for optimal cross selection.”

Matt Baseggio
Graduate student, Cornell University
Michael Gore Lab

“Large proportions of the US population do not meet the daily-recommended intake of several vitamins and nutrients. My research is trying to improve the nutritional quality of sweet corn—the third most consumed vegetable in the US. I conducted genome-wide association studies to identify genes and favorable alleles responsible for quantitative variation of kernel carotenoid (provitamin A, lutein, zeaxanthin), tocochromanol (vitamin E), and nutrient (iron and zinc) levels in a sweet corn diversity panel. I am also developing and validating marker-based prediction models to convert locally adapted sweet corn germplasm to dark orange kernel with high vitamin and nutrient content.”

Anthony Findley
MD/PhD student, Wayne State University
Roger Pique-Regi and Francesca Luca Labs

“My research focuses on gene regulation in varying environmental contexts and cell types. I integrate gene expression and chromatin accessibility data from cells treated with a variety of hormones, environmental contaminants, drugs, and metals to identify regulatory elements which modulate cellular response to each condition. I am particularly interested in linking these in vitro exposures with complex traits and understanding how genetic variation in environmentally responsive regulatory elements can be used to predict disease susceptibility.”

Margaret Krause
Graduate student, Cornell University
Mark Sorrells and Michael Gore Labs

“My research focuses on the integration of high-throughput phenotyping and genomic selection in plant breeding. Advances in remote sensing have enabled plant breeders and geneticists to collect an extensive amount of phenotypic information on large numbers of individuals throughout their growth and development. Integrating these traits into genomic selection has the potential to increase the rate of genetic gain in crop plants.”

Jhonathan Pedroso Rigal dos Santos
Graduate student, University of São Paulo and Cornell University
Michael Gore Lab

“The ability to connect information between traits over time allow Bayesian networks to offer a powerful probabilistic framework to design genomic prediction models. In our research, we phenotyped (plant height time series, biomass) and genotyped (100,435 SNPs) a diverse panel of 869 sorghum lines. We developed the models Bayesian Network (BN), Pleiotropic Bayesian Network (PBN), and Dynamic Bayesian Network (DBN). For benchmarking, we used multivariate GBLUP models. The DBN model approached the same accuracy as the reference model and allowed to compute probabilistic indexes to identify optimal time points before the end of the season for earlier indirect selection.”

Fabio Morgante
Postdoc, University of Chicago
Yang Li and Matthew Stephens Labs

“I am interested in statistical and quantitative genetics in a variety of species, with a special focus on prediction of complex traits. While my initial work involved analyzing livestock data, I quickly transitioned to using Drosophila melanogaster as a model system to develop statistical models and analysis strategies to predict complex traits more accurately, by leveraging multiple layers of data (e.g., genomic, transcriptomic, metabolomic). Recently, I have become interested in human genetics and have been working on a method that exploits the sharing of eQTL effects among different tissues to increase the prediction accuracy of expression levels from genotype data.”

Ivone de Bem Oliveira
Postdoc, University of Florida
Patricio Muñoz Lab

“My research has been focused on the intersection between breeding and genomics, particularly in developing solutions to improve the selection process for polyploid breeding. Our pioneering research has proven the feasibility of genomic prediction for blueberry, enabling reductions in breeding cycle times and increasing genetic gain. Now we are optimizing the relationship between genotyping cost and model accuracy for an economically feasible application of genome prediction for blueberry. I am evaluating the effect of number of markers, sequencing depth, and training population on phenotype prediction. The benefits and pipeline described in our studies can be applied to other polyploid species.”

Blaise Ratcliffe
Postdoc, University of British Columbia
Yousry El-Kassaby Lab

“My research focuses on the integration and use of genomic information in conifer tree improvement programs. Genomic selection tools have the potential to accelerate rates of genetic gain for complex, quantitative traits through early prediction of phenotypes and increased selection intensity. These new tools enable breeding programs to respond rapidly to the changing market demands of forest products as well as emerging abiotic and biotic threats.”

Palle Duun Rohde
Postdoc, Center for Quantitative Genetics and Genomics at Aarhus University
Peter Sorensen Lab

“My work focuses on the accurate prediction of individual disease risk or the response to medical treatment is important for the development of precision medicine. To successfully advance in precision medicine a better understanding of the genetic architecture of human complex traits and disease is required. In my research, I focus on statistical genetic methods for integration of different types of data to achieve a better understanding of the genetic basis of complex traits and diseases. Currently, my work is focused on how to leverage multi-layered phenotypes and multi-layered molecular data to improve current prediction models, in particular with respect to treatment response.”

Nicholas Schreck
Postdoc, University of Mannheim
Martin Schlather Lab

“I focus on the theoretical analysis of mixed linear models with special focus to the estimation of the additive genomic variance. I also investigate coefficients of determination in mixed linear models with the aim of efficient variable selection for high-dimensional genomic data sets.”

Gregory Way
Postdoc, Broad Institute of MIT and Harvard
Anne Carpenter Lab

“My PhD focused on developing supervised and unsupervised machine learning approaches to extract knowledge from large publicly available gene expression data sets. I have developed approaches to isolate gene expression signatures in tumors including identifying Ras pathway activation and TP53 inactivation signatures. In my postdoc, I will focus on extracting knowledge from large biomedical imaging data sets. My goals include developing methods that measure subtle responses to drug treatments in cell lines and to integrate imaging and gene expression data to provide additional views towards solving difficult biomedical problems.”

Yvonne Wientjes
Postdoc, Wageningen University
Research Animal Breeding and Genomics

“Genomic selection has revolutionized artificial selection in livestock populations. Compared to classical selection, genomic selection is more accurate, focusses more on genes with large effects and ignores rare genes or with small effects. Therefore, genomic selection has likely increased the change in allele frequencies of genes over generations and may have changed the effects of those genes when non-additive effects are present. I aim to investigate how fast genomic selection methods change the genetic architecture of traits, i.e. the allele frequencies and effects underlying the trait. This will provide information on whether current selection methods limit the potential for long-term genetic improvement.”

Alencar Xavier
Research Scientist, Corteva Agrisciences and Purdue University
David Habier Lab

“My work on statistical genetics is focused on genomic-assisted breeding with emphasis on theoretical and computational aspects of data-driven plant breeding, such as modeling, prediction and selection using various sources of information. My research regards the development and implementation of new quantitative methods using mixed models, Bayesian methods and machine learning, along with high-performance computing.”

Robert Baker
Assistant Professor, Miami University

“I study organismal evolution of plant form and function. To do so, I examine the connection between genotypes and phenotypes throughout development and across environments at the intraspecific level. My research integrates quantitative genetics, genomics, transcriptomics, anatomy, morphology, and physiology in plants from natural, model, and crop systems. I use these data in part to predict novel, non-linear developmental phenotypes based on genotypes. My work has implications for understanding natural biodiversity, conservation and restoration, and improving agricultural sustainability.”

Helena Oakey
Senior Research Fellow, University of Adelaide
The Biometry Hub, Olena Kravchuk Lab

“My research interests cover the development and improvement of statistical genetic methodology, including genomic selection and association mapping to account for factors unique to agronomic trials such as replication, multiple phases (laboratory and field), treatments and environment.”

A landmark paper published in GENETICS founded the field of genomic prediction before the requisite technology was available.

When a new technology is developed, it can allow scientists to make great strides in addressing longstanding questions. Occasionally, however, researchers think so critically about a knowledge gap in their field that they’re able to propose a new methodology that anticipates the technology needed to make it a reality.

This is precisely what Theo Meuwissen, Ben Hayes, and Mike Goddard accomplished with their 2001 paper Prediction of Total Genetic Value Using Genome-Wide Dense Marker Maps. In it, they laid out a framework for predicting breeding values from genome-wide marker information, using simulated data to compare different approaches. The catch? There wasn’t a way to do what they were proposing—the technology didn’t exist yet.

Despite this seemingly major drawback, the authors were able to successfully use theory and simulated data to propose methods that would one day prove to revolutionize animal breeding strategies.

“In retrospect, the paper was a bit of a thought piece,” says Hayes. “Imagine if we could do this: what would it look like?”

The central goal of selective animal and plant breeding is increasing the genetic gain—that is, enhanced performance—of economically important traits. This was classically achieved by meticulously recording individuals’ phenotypic information in a population and using these records to estimate breeding values and select the best breeders for establishing the next generation. As the genomic era began to bloom toward the end of the 20th century, researchers began to incorporate genotype data into their selection strategies.

“The prevalent attitude was to try and map individual quantitative trait loci (QTLs) and then incorporate them into decisions about selection of animals,” according to Goddard.

But most of the traits in question were not associated with a small number of genes or markers, as originally anticipated. Instead, the relevant traits were likely controlled by many genes of small effects—hundreds or even thousands of genes, in fact. Existing methods were geared toward mutations of large effect, which the field was discovering weren’t likely to be found.

As the complexity of the genomic architecture underlying these traits was becoming clearer, genotyping technologies were becoming more advanced.

“It had been predicted that we would get dense marker data, but we didn’t know what to do with it. We were trying to figure out what to do if we were able to get dense marker data in a cost-efficient way,” says Meuwissen.

They explored a genome-wide approach to predict breeding values without mapping specific QTLs. They needed a high density of markers across the genome for this type of approach to work, but since that kind of real data didn’t exist yet, they simulated a genome and marker set and tested a number of statistical methods. After comparing linear regression, Best Linear Unbiased Prediction (BLUP), and multiple Bayesian methods (termed BayesA and BayesB), they concluded that “selection on genetic values predicted from markers could substantially increase the rate of genetic gain in animals and plants, especially if combined with reproductive techniques to shorten the generation interval.”

They published their work in GENETICS, noting presciently that “the advent of DNA chip technology may make genotyping of many animals for many of these markers feasible (and perhaps even cost effective).” But since SNP chips weren’t yet in the hands of researchers, the paper didn’t spark an immediate revolution in quantitative genetics or animal breeding. Meuwissen, Hayes, and Goddard had founded the field of genomic selection (also termed genome-wide prediction), but the full potential of their findings wouldn’t be realized for a number of years.
“The paper really sat in the cupboard until the technological advance came along,” says Hayes.

Thankfully, they didn’t have to wait too long: by the end of the decade, SNP chips—which allow simultaneous genotyping of thousands of markers—were available for major livestock species. And with the availability of SNP chips came an explosion of interest in the paper that founded genomic selection.

Citations to Meuwissen et al. (2001) according to PubMed and Google Scholar. From de Koning (2016).

In the nearly two decades since, the field has grown and changed in a variety of ways. For one, genotyping technology has continued to improve.

“It started off being a relatively small number of SNPs (~10,000 on the first bovine chip), and now you can get 600,000. SNP tech came onstream and rapidly advanced,” notes Goddard.

Additionally, these methodologies have also been applied more widely than livestock breeding—most notably to plant breeding and to human genetic studies of disease risk prediction. For more insight into the similarities and differences in how the methods are applied in different settings, see the new review published this month in GENETICS by Naomi Wray and colleagues.

What’s next for genomic prediction?

Researchers are still working on the best way to use whole genome sequencing (WGS) data instead of SNP chip data—though it’s now easier and cheaper than ever to sequence entire genomes, there hasn’t been much advantage to using WGS data over SNP data to date.

There are also challenges related to applying genomic prediction across breeds.

“Doing genomic prediction across breeds really doesn’t work well at the moment,” explains Hayes. “This is a problem because, in some breeds, it’s cost prohibitive to build the populations needed to drive genomic selection. There’s a lot of work going on about borrowing information across breeds.”

And as genomic prediction is being implemented widely and in many different species, it’s important for breeders to keep an eye on genomic diversity within their populations.

“We’re getting increasingly effective tools, but if we run out of diversity, we won’t be able to maintain the selection response we see today into the future,” notes Meuwissen.

Through the intervening years, the methods laid out in the 2001 paper have stood the test of time, with BayesB remaining at the forefront of genomic prediction. The field continues to grow and develop, moving into new species and honing the technologies—goals aided by the Genomic Prediction series launched in 2012 at the GSA Journals. Since then, GENETICS and G3 have collected an exciting body of work, encouraging the exploration of methods and the sharing of data to advance the field.

Genomic prediction is a striking demonstration of how science needn’t be limited by existing technology. In some cases, theoretical advances can even predict the future and help us make the most of technological advance.

CITATIONS

Prediction of Total Genetic Value Using Genome-Wide Dense Marker Maps
T. H. E. Meuwissen, B. J. Hayes and M. E. Goddard
GENETICS April 2001, 157 (4): 1819-1829.
http://www.genetics.org/content/157/4/1819

Meuwissen et al. on Genomic Selection
Dirk-Jan de Koning
GENETICS May 2016, 203(1): 5-7.
https://doi.org/10.1534/genetics.116.189795
http://www.genetics.org/content/203/1/5

Machine learning and access to ever-expanding databases improves genomic prediction of human traits.

In theory, a scientist could predict your height using just your genome sequence. In practice, though, this is still the stuff of science fiction. It’s not only your genes that affect height—environment also plays a role—but the larger problem is that height is affected by tens of thousands of individual genetic variations. This is also true of other complex traits, such as susceptibility to particular diseases. To get closer to accurate genomic prediction of human traits, geneticists are using new approaches to harness the vast amounts of sequence data becoming available. In GENETICS, Lello et al. describe a machine learning approach to the problem that allowed them to make predictions within a few centimeters of reality.

“To me, genomic prediction is the actual decoding of the genome,” says senior author Stephen Hsu from Michigan State University. A theoretical physicist by training, Hsu explains that his lab became interested in the problem of genomic prediction several years ago as the cost of genotyping continued to drop and more datasets became available. They had previously argued that they could predict complex traits, like height, if they only had enough data.The release of nearly 500,000 UK Biobank genotypes allowed them an opportunity to test this hypothesis.

A genomic prediction approach is quite different from the more familiar genome-wide association study (GWAS). GWAS methods test each SNP one at a time, looking for statistically significant contributions to the phenotype. In contrast, genomic prediction makes use of all SNPs at once in trying to build the best possible predictors.

The authors took the Biobank genotype and phenotype data and used a type of regression to identify the combination of SNPs that, taken together, best correlate with the trait of interest. Since only a subset of SNPs influence each trait—even the thousands of loci that control height are only a tiny fraction of the total number of SNPs identified —they also introduced a penalization factor that prevents the model from including unneeded SNPs. They were essentially trying to solve an optimization problem: identify the fewest number of variables (i.e. SNPs) that will allow for the best prediction about the outcome (i.e. trait).

Having generated their algorithm, the authors then put it to the test. They constructed models for height, heel bone density, and educational attainment, and they found that their algorithm worked well, particularly for height. For example, it produced a nearly 0.65 correlation with actual height, and predicted heights were usually within a few centimeters of actual heights. “Our predictor actually captures almost all the heritability that we could expect to find,” says Hsu.

With enough data, Hsu believes, accurate genomic prediction for complex traits will no longer be sci-fi. As more and more genotypes are obtained, Hsu predicts that this kind of prediction could be applied for most traits in as little as five years.

CITATION:

Accurate Genomic Prediction of Human Height

Louis Lello, Steven G. Avery, Laurent Tellier, Ana I. Vazquez, Gustavo de los Campos, Stephen D. H. Hsu

Genetics October 2018 210: 477-497; https://doi.org/10.1534/genetics.118.301267

http://www.genetics.org/content/210/2/477

Study of huge Ancestry.com pedigree suggests assortative mating may have inflated previous estimates of life span heritability.

Although long life tends to run in families, genetics has far less influence on life span than previously estimated, according to a new analysis published in GENETICS.  Ruby et al. used a data set of over 400 million historical persons obtained from public pedigrees on Ancestry.com to estimate the heritability of life span, finding it to be well below 10%.

“We can potentially learn many things about the biology of aging from human genetics, but if the heritability of life span is low it tempers our expectations about what types of things we can learn and how easy it will be,” says lead author Graham Ruby (Calico Life Sciences). “It helps contextualize the questions that scientists studying aging can effectively ask.”

Calico Life Sciences is a research and development company whose mission is to understand the fundamental science of aging. So how did Calico get involved with Ancestry, the online genealogy resource?

“We wanted to get a sense for the contribution of genetics to life span, and that’s something you can study using pedigrees,” says Ruby. With millions of members, Ancestry has no shortage of pedigrees.

Fortuitously, researchers at Calico and Ancestry were connected from their time in academic basic research. Calico’s Chief Scientific Officer David Botstein and Ancestry’s Chief Scientific Officer Catherine Ball (senior author on the GENETICS paper) both have backgrounds in yeast research. They were involved in the Saccharomyces Genome Database project during their times at Stanford University and published a number of papers together.

So researchers from both companies teamed up to use publicly available pedigree data from Ancestry.com to approach the problem of figuring out the genetic contributions to human longevity.

“Partnering with Ancestry allowed this new study to gain deeper insights by using a much larger data set than any previous studies of longevity,” says Ball.

The heritability of life span has been well investigated in the literature, with previous estimates ranging around 15-30%.

But some of these studies found that it wasn’t just blood relatives who shared similar life spans—so did spouses. This suggested that the heritability estimates might have been confounded by shared environments or assortative mating (the tendency to choose mates who have similar traits to ourselves).

The new study had the power to investigate these possibilities in more detail because of the large size and high quality of the data. The data set, called the SAP for “set of aggregated and anonymized pedigrees,” was constructed from Ancestry.com pedigrees with the help of source references like birth certificates.

Starting from 54 million subscriber-generated public family trees representing six billion ancestors, Ancestry removed redundant entries and those from people who were still living, stitching the remaining pedigrees together. Before sharing the data with the Calico research team, Ancestry stripped away all identifiable information from the pedigrees, leaving only the year of birth, year of death, place of birth (to the resolution of state within the US and country outside the US), and familial connections that make up the tree structure itself.

The SAP included almost 500 million individuals (with a single pedigree accounting for over 400 million people), largely Americans of European descent, each connected to another by either a parent-child or a spouse-spouse relationship. The scale of the data allowed the researchers to get accurate heritability estimates across different contexts; they could stratify the data by birth cohort or by sex or by other variables without losing the power needed for their analyses. They employed structural equation modeling—a technique that hasn’t often been applied to this problem due to the amount of data required for it to be productive—to calculate life span correlations and heritability across the giant pedigree.

Running the numbers, the team initially found heritability estimates to be between 15-30%—similar to the reported literature.

“But then I did correlations between first cousins-in-law, and their life spans didn’t correlate as much—but it was close,” says Ruby.

When a trait correlates between in-laws similarly to blood relatives, that can mean that something besides genetics is being shared across households. The term heritability describes the proportion of trait variability in a population that can be attributed to genetic differences. But genetics aren’t the only thing that can be passed down between generations: sociocultural factors can also influence certain traits, and these too can be inherited. The combination of genetic heritability and sociocultural heritability is the total transferred variance, that is, the total amount of variability in a trait that can be explained by inheritance.

“The comparison between in-law relatives is something that hadn’t been as thoroughly explored in the prior literature. With this large dataset, we could look at in-law relatives at a large scale and be confident that they weren’t that far off from blood relatives,” says Ruby.

The scale of the data allowed Ruby and colleagues to look not only at siblings-in-law and first cousins-in-law but also to examine correlation in both types of co-siblings-in-law (your sibling’s spouse’s sibling or your spouse’s sibling’s spouse). None of these relationship types generally share household environments, and yet their life spans showed correlation.

If they don’t share genetic information and they don’t share household environment, what accounts for the similarity in life span between individuals within these relationship types? Going back to their impressive dataset, the researchers were able to perform analyses that detected assortative mating.

“What assortative mating means here is that the factors that are important for life span tend to be very similar between mates,” says Ruby. In other words, people tend to select partners with traits like their own—in this case, how long they live.

Of course, you can’t easily guess the longevity of a potential mate. “Generally, people get married before either one of them has died,” jokes Ruby. Because you can’t tell someone’s life span in advance, assortative mating in humans must be based on other characteristics.

The basis of this mate choice could be genetic or sociocultural—or both. For a non-genetic example, if income influences life span, and wealthy people tend to marry other wealthy people, that would lead to correlated longevity. The same would occur for traits more controlled by genetics: if, for example, tall people prefer tall spouses, and height is correlated in some way with how long you live, this would also inflate estimates of life span heritability.

By correcting for these effects of assortative mating, the new analysis found life span heritability is likely no more than seven percent, perhaps even lower.

The upshot? How long you live has less to do with your genes than you might think.

CITATION

Estimates of the Heritability of Human Longevity Are Substantially Inflated due to Assortative Mating
J. Graham Ruby, Kevin M. Wright, Kristin A. Rand, Amir Kermany, Keith Noto, Don Curtis, Neal Varner, Daniel Garrigan, Dmitri Slinkov, Ilya Dorfman, Julie M. Granka, Jake Byrnes, Natalie Myres, and Catherine Ball.
GENETICS November 2018. 210(3): 1109-1124.
http://www.genetics.org/content/210/3/1109
DOI: 10.1534/genetics.118.301613

We are delighted to announce the finalists for the James F. Crow Early Career Researcher Award! All finalists will speak at the Crow Award session of the Population, Evolutionary, and Quantitative Genetics Conference on May 14, 2018 in Madison, Wisconsin.

The Award honors the legacy of James F. Crow, whose contributions to the field of genetics were impactful and innumerable. Learn more about Crow and the award.

Crow Award finalists: Top row, left-right: Jeremy Berg, Alison Feder, Amy Goldberg; bottom row, left-right: Emily Josephs, Emily Moore, Katherine Xue

Jeremy Berg

Columbia University

I have always been interested in how things work. As a child, this manifested first as an interest in understanding how various pieces of machinery worked, and later in the laws of physics and cosmology. As an undergraduate student at the University of Wisconsin (where I was fortunate enough to have Crow as a guest lecturer a few years before his death), my interests turned toward the biological, and particularly to evolution. It seems natural then that I was drawn to population genetics, as it is the science which deals on a mechanistic level with understanding how the evolutionary process works, and I chose to pursue a PhD in this area at UC Davis.

In particular, my work focuses on understanding how evolutionary processes played out over the past ~200,000 years have contributed to the generation of diversity within the human population. Sometimes, this takes the form of understanding recent adaptations that have evolved within human populations, which gives us an insight into the challenges that our ancestors faced and the kind of traits that helped them survive and prosper. Another branch of my work focused more on the role of evolution in genetic disease, and in particular understanding why certain genetic diseases persist in the population despite their obvious fitness costs, and what forces are respsonsible for shaping their underlying genetic architectures.

Go to Jeremy Berg’s presentation abstract.

Alison F. Feder

Stanford University

I loved both math and life sciences classes throughout school, so when I started college, I sought out a research experience in quantitative biology. This search led me to Warren Ewens and Joshua Plotkin, who showed me how quantitative approaches could be especially instructive in understanding evolutionary biology. For example, when a population changes over time, can we use statistics to determine whether chance or selection drives those changes? This research question, which I pursued in Dr. Plotkin’s lab as an undergraduate, seeded a longstanding interest in understanding adaptation quantitatively.

I want to understand how and when populations adapt in the face of daunting evolutionary odds. Whereas most empirical and theoretical work has considered populations evolving to a single challenge, in nature, these challenges often are multiple and sometimes orthogonal. For example, we currently treat HIV with combinations of three drugs that attack different stages of the viral life cycle. This treatment strategy has substantially reduced but, importantly, not eliminated, the evolution of drug resistance within patients. How does adaptation still occur under such improbable conditions? I use HIV as a case study to understand evolution to selective pressures of varying complexity, including those structured in physical space. Many natural and man-made environments present similarly difficult challenges to species in nature. What separates the conditions that allow some species to persist via adaptation but drive others to extinction? Long-term, I hope to address such questions in a way that both improves human health and advances our fundamental understanding of evolution.

Go to Alison Feder’s presentation abstract.

Amy Goldberg

University of California, Berkeley

The recent availability of large high-coverage genetic datasets, combined with careful ecological sampling, has demonstrated the outsized impact that recent evolutionary history—the last tens of generations—can have on the genetic and phenotypic variation of a population. I develop quantitative methods to elucidate recent population histories, and interpret these histories in the context of demographic, cultural, and environmental pressures. Starting my scientific career with archaeological fields schools, my strong interest in mathematics and a desire to quantitatively test hypotheses drew me to population genetics.

Admixture, or hybridization, is one of the fastest evolutionary processes to radically change the composition of a population. Admixed populations are formed through the exchange of individuals from two or more previously isolated populations. Acting as a natural experiment, admixed populations offer insight into adaptations of their parental populations, and are ubiquitous throughout animal and plant populations. Multiple processes such as inbreeding avoidance and phenotypic mate preferences direct how parental populations mix. Despite this complexity, previous methods often considered admixed populations as simple instantaneous combinations of their sources. Instead, my work takes a mechanistic approach, deriving flexible sets of mathematical models to consider the admixture process, which often varies in intensity geographically or over time. Importantly, my theoretical results demonstrate that previous work—which does not consider sex bias, nonrandom mating, or the mechanistic process—misestimates population history parameters, such as the timing and intensity of migration. My work also links admixture dynamics with classic population genetic models of breeding and assortative mating by genotype.

Go to Amy Goldberg’s presentation abstract.

Emily B. Josephs

University of California, Davis

I was always interested in being some sort of scientist and experiences volunteering at a small aquarium, along with taking an conservation biology class in high school, fostered that interest. However, it was in my first evolutionary biology class, when I discovered population genetics, that I got excited about evolutionary biology. This enthusiasm led me to a research assistant position in an evolutionary ecology lab, where I got to participate in the often-tedious-but-occasionally-exciting business of actual science. After graduation, I worked as a technician studying speciation in wild tomatoes and rapid adaptation in Trinidadian guppies before starting graduate school at the University of Toronto. I spent my PhD trying to disentangle how selection and drive shape genetic variation in the weedy plant species Capsella grandiflora. I found that negative selection against new mutations and positive selection for new mutations shape broad patterns of genomic variation and that much of the genomic variation that actually affects gene expression is under negative selection. Now, as a postdoctoral researcher at the University of California, Davis, I am continuing to investigate the evolutionary forces shaping variation by studying if and how local adaptation contributes to quantitative trait variation in domesticated maize. My goal is to not only answer these questions in maize but to also develop methods that will be useful for investigating additional species and, overall, I hope that my work linking patterns of genomic variation with trait variation will contribute to a better understanding of how evolution shapes the world around us.

See Emily Josephs’ presentation abstract.

Emily C. Moore

North Carolina State University

One of my overarching research questions is ‘how do changes to the genome allow for the varied patterns of behavior we see reflected across the animal kingdom?’ I was inspired to start exploring this idea during my undergraduate study of biology and philosophy of mind, and I hope to continue to ask questions like these through the rest of an academic career. I have always been curious about how biological systems work, though it wasn’t until my early twenties that I recognized that a career in scientific research would be more engaging than a career in medicine—where I could contribute to the field of knowledge rather than merely absorb it. One of the focuses of my doctoral work has been to understand the genetic basis of behavioral differences between ecologically-varied cichlid fish species. I have been fortunate to be able to identify an interesting phenotype (an exploratory behavior), explore the genetic architecture of the behavior, and ultimately identify and evaluate a specific genetic variant within the time span of a PhD. The beauty of asking these questions in an ecologically and evolutionarily interesting vertebrate model organism is that our findings are relevant to the fields of evolutionary ecology and vertebrate neurobiology. The more we understand about how natural variation in the vertebrate genome shapes the development and function of the brain, the better insight we can have into how behavioral patterns evolve, and how disruption to neurogenetic pathways can lead to brain and behavioral dysfunction.

Katherine Xue

University of Washington

I never liked biology before I started learning about evolution. As a kid, I was fascinated by the order and rigor of math, but biology seemed like a chaotic mess. It wasn’t until high school that a teacher suggested I read books by Stephen Jay Gould and Richard Dawkins. Before long, I was hooked. I wanted to understand how evolutionary principles could unify and explain what Darwin called life’s “endless forms most beautiful.”

I started combining my quantitative leanings with my new love of biology. As an undergraduate, I bred tetraploid plants to understand how meiosis adapts to the tangled complications of having extra chromosomes. After graduating, I worked for a year as a science writer, inspired by the writers who first brought me to evolutionary biology.

As a graduate student, I study the lightning-fast evolution of influenza. Flu viruses change incredibly rapidly, requiring constant monitoring and vaccine updates. Recent advances in genome sequencing make it possible to track this evolution at high resolution. My work has shown that flu viruses can rapidly and repeatedly evolve to cooperate with one another in laboratory settings. I’ve also used new sequencing technologies to zoom in on how flu evolves in individual people while they’re sick. Surprisingly, I found that flu viruses can evolve in infected people over time in ways that mirror global evolution, which helps us understand better how flu evolution starts. I find it exciting and inspiring that evolutionary principles can shed light on important problems affecting human health.

Go to Katherine Xue’s presentation abstract.

Abstracts