GACT: Research

New project: Quantitative Genetics of Hybrid Yeasts

A newly funded BBSRC project to exploit diversity and biotechnological uses of hybrids by overcoming sterility. A collaboration between the Universities of Leicester, Manchester and Nottingham and SABMiller. This is complemented by an iCASE studentship on brewing yeast hybrids also in collaboration with SABMiller. The primary goal is to open up yeast hybrids, important in many industrial fermentations, to genetic analysis by overcoming the hybrid sterility. We want to both explore all the possible genetic variation and combinations from the 7 extant species and from some existing hybrids in use, and generate new strains with properties of interest to different fermentation situations.

Selection for improved characteristics of crops and animals over the millennia of modern human history has benefited from the ability to breed. Artificial selection for such improvement through selective breeding feeds the world as well as gives us the wide variety of dog breeds, faster race horses, etc. Without the ability to breed there is limited variation on which to base improvements. Many plants, including various crops of great importance such as wheat, originated from the hybridization of two or more closely related species. These initially were sterile and dead ends in terms of evolution but overcame their sterility by duplicating their genomes. Variation could still be limited in these as the fertile derivatives may have occurred only once in history resulting in variation being fixed at that point and accumulated over time via mutations. A lot of effort goes into crop improvement by introgressing chromosomes from parental species of the hybrids to bring in new variation. In animals it is more difficult to overcome sterility and so working animals such as mules can only really be improved by breeding in the horse and donkey parents and hoping for improved characteristics in each new mules created by mating the two parents. This is a hit or miss approach and not very efficient.

In fermentation uses with yeast there are many hybrids that are used, the most famous being the lager yeast Saccharomyces carlsbergensis. There are also some used in the wine industry and others are found in nature. These are sterile hybrids which preclude improvement by breeding. In this project we overcome the sterility barrier by duplicating the genomes of several new and existing hybrids, first to determine the genetics of particular characteristics, like why does lager yeast ferment better at cold temperatures, and then to create new diverse hybrids with improved or even new characteristics. We will generate a large number of new hybrids and explore their characteristics, isolating useful strains for use in brewing and wine making as well as industrial production strains. We will also learn about the biology of hybrids and how their two genomes interact. Finally we will answer the question ‘Can mules evolve?’

Complex Traits and Missing Heritability

One of the big problems in determining the underlying genetic causes of complex traits is called 'missing heritability'. Most traits including many human diseases have many genetic factors underlying their expression and are not inherited in a simple Mendelian fashion. Variants at many individual regions can be associated with a trait but in general when taken as a whole they explain only a fraction of the heritable contribution to the trait. This is particularly relevant to many human diseases. The heritability is not really missing but is masked by complex interactions between different loci as well as with the environment. In human GWAS (genome wide association studies) studies, many QTLs (quantitative trait locus) may be identified as having an association with the expression of a disease state but in general there is no statistical power remaining for detecting any two or more locus interactions. Recent studies in yeast are making headway into cracking the problem of detective non-additive or epistatic interactions among the genes involved in a trait.

Observations from Yeast

In the past decade a few classical breeding experiments in yeast have been done to determine the genetic factors involved in some traits that were not simply inherited such as sporulation efficiency and heat tolerance. These experiments generally involved the main lab strain crossed to one other yeast strain from a different environment. Much has been learned from these studies about the underlying genetics of these traits but these studies also revealed that determining the underlying genetics is not simple. More recently a survey of genome sequences along with the measure of multiple phenotypes in many isolates of yeast from diverse locations and sources has revealed a great deal of phenotypic variation which correlates with genetic relationships (Liti et al., 2009. Population genomics of domestic and wild yeasts. Nature 458: 337-341). As the majority of phenotypes measured are quantitative and the correlation was good, we were confident that appropriate crosses using different isolates would be informative for determining the underlying genetic architecture of these complex traits. Through a series of pairwise crosses (Cubillos et al., 2011. Assessing the complex architecture of polygenic traits in diverged yeast populations. Molecular Ecology 20: 1401–1413) we were able to demonstrate that underlying QTLs could be identified and confirmed for many traits of interest. Several observations point to the consequences of complexity and problems still to be solved in order to crack the interactions that result in apparent 'missing heritability'. Liti and Louis (2012. Advances in Quantitative Trait Analysis in Yeast. PLoS Genetics 8(8): e1002912) reviews our latest thoughts on quantitative genetics in yeast.

  • Transgressive variation - in virtually every cross we see phenotypic variation in the offspring more extreme than either parent.
  • Antagonistic variation - up to 1/3 of the QTLs have their phenotypic effects in the opposite direction than expressed in the parent.
  • Subtelomeric QTLs - 25% of QTLs for traits overall map beyond the last unique markers though the unmapped/unassembled subtelomeric regions comprise only 8% of the genome.
  • Interactions with undetected individual QTLs - we observed a 2-locus genotype associated with a trait where neither locus was detected as having an association.
  • Linked QTLs of mixed effect - high resolution mapping of QTLs indicates that they tend to be clustered physically and are a mixture of antagonistic and expected direction QTLs.

iQTLs and High Resolution Mapping

A major advance in QTL analysis in yeast came through pooling segregants of F1 hybrids and comparing the sequence of a pool of selected progeny to the unselected pool. This X-QTL method developed by the Kruglyak group (Nature 464:1039-1042, 2010) initially still had the problem of low resolution due to linkage though increasing the number of segregants selected helps greatly (Nature 494:234-237, 2013). We have taken this further by breaking up linkage associations using advanced inter-cross lines. Our iQTL approach lead to the identification of the majority of QTLs for a given trait down to the gene and some down to the causal nucleotide or QTN (Parts et al.,  2011 Revealing the genetic structure of a trait by sequencing a population under selection. Genome Research 21: 1131-1138)

The next steps at GACT

The initial studies within GACT will include the following:

  • Solving subtelomeric assembly
  • Directly measuring two- and multi-locus interactions
  • Testing one corollary of co-adapted complexes
  • Generate a population of yeast for selecting traits of interest

Share this page:

Contact GACT

To contact the Centre for Genetic Architecture of Complex Traits, please email Professor Ed Louis

The postal address is:

GACT, Department of Genetics, University of Leicester, Leicester LE1 7RH, UK

Phone: (+44) 0116 229 7813

Contact Details

College of Life Sciences
University of Leicester
Maurice Shock Building
University Road