In genomic prediction statistical models are used to predict phenotypes from genetic marker data (such as SNPs). Genomic prediction has proven to be superior to pedigree based inference, but against expectations the use of more markers did not greatly improve genomic prediction. The aim of this project is to improve genomic prediction by implementing biological information (such as the structure of the genome) and multivariate statistics. This project is part of TOPBREED , a research project of the Breed4Food consortium .

In the traditional framework of evolution, mutations are regarded as the ultimate source of genetic variation and natural selection as the filtering process determining which variants are conserved for future generations and which are not. Effects of the environment as well as developmental processes on the way genotypes translate to phenotypes are regarded random and not affecting evolution. However, the developmental system organisms are subject to is also shaped by evolution, therefore the way genes and the environment affect the phenotype are coordinated and can be directional and even functional. The evolutionary trajectory a species takes is therefore affected by development. This developmental bias is often marginalized as anecdotal. In this project we will use meta-analysis to investigate how widespread developmental bias is, whether cryptic genetic variation — which is released in novel environments — is non-random and how phenotypic plasticity plays a role in these processes. This project is part of the Extended Evolutionary Synthesis research program.

The amounts of genomic and phenotypic data collected in ecology and evolution are surging. The collection of more and new types of data require new analytical techniques; generalized linear mixed models do not always offer the right tools and therefore more flexibility is needed. Among other benefits, Bayesian statistical techniques offer tremendous flexibility of model structure and the integration of uncertainty. I use Stan, a general-purpose Bayesian inference library and language, which utilizes novel algorithms to efficiently deal with highly dependent data structures. I develop and use novel model structures to analyze high dimensional dependent data structures to for instance infer extra-genetic or indirect genetic effects.

When exposed to cyanobacteria, daphnids are known to build up resistance to microcystin — a toxin produced by cyanobacteria — over (clonal) generations. This could be a positive maternal effect in which mothers prepare their offspring to the environment they will encounter, by for instance adding particular proteins to the eggs. However, a toxic environment also affects the growth of mothers, which ultimately affects the amount of resources they can provide to their offspring and therefore the offspring’s fitness. In this project we investigate the interplay between those two (positive and negative) maternal effects. We investigate the effect of different levels of toxicity on different genotypes and life phases. As part of this project we also test whether DNA methylation can act as a mechanism of inheritance for this resistance to microcystin.

Currently I am working on genetics and social structure in tit species. A particular dynamic relationship is present between genes and social structure, because gene flow is strongly influenced by the social structure of a species while genes have the ability to dictate social behaviour. We collect data on the social structure of various passerines (Great tits, Blue tits, Marsh tits and Nuthatches) by registering visits to feeding stations at the individual level with radio frequency identification tags. We translate the social structure into social networks to quantify "social phenotypes". For Great tits a SNP-chip was developed and for two years most breeding pairs were genotyped. We also have pedigree data for both Great tits and Blue tits, which enables us to use "animal models" to infer genetic effects. Combining the social network data and the genotypes enables us to address questions regarding genetics and social structure. I focus on investigating the effect of social behaviour on the spatial distribution of genotypes, mate choice and genetics, quantitative genetics (both pedigree and marker based) of social behaviour and the evolutionary consequences of social behaviour. This project is part of a large project on the role of social processes in evolutionary ecology funded by an ERC grant awarded to Prof Ben Sheldon.

Sex allocation theory is successful in predicting sex ratio variation in some taxa, but often not in birds and mammals. An important reason may be that most theoretical models do not account for the complexities of avian and mammalian life-history. To reveal the adaptive significance of sex allocation to avian life-history, I conducted an experimental study on great tits (Parus major). I investigated the effect of manipulated brood sex ratios on long-term fitness benefits. Offspring that was raised in broods with sex ratios at parity had the highest fecundity. Since offspring recruitment, parental survival and parental future fecundity were unaffected, there was stabilizing selection for producing brood sex ratios at parity. To gain insight in the underlying mechanism I focused on different aspects of the offspring�s first year. Brood sex ratio did not affect fledging behaviour, but there were some effects on social behaviour in winter, which did not correlate to fitness. As adults, individuals raised as the rare sex in a brood had larger tarsi than individuals of the abundant sex. This relationship was neither present in the nestling phase nor caused by selective disappearance. This suggests a sex-specific directional effect of brood sex ratio on the late development of the offspring, which potentially links brood sex ratio to fecundity. Stabilizing selection for brood sex ratios at parity might provide an explanation why sex allocation in birds and mammals is often subtle. Higher benefits for equal brood sex ratios counteract selection for facultative sex allocation to extreme values.

One of the most dramatic changes a cavity nesting bird will ever experience in life is leaving the safety of the nest it was born in and start exploring the outside world. In this project we try to understand of the social dynamics around this moment called fledgling.

Bewick's swans (Cygnus columbianus bewickii) breed in Siberia and winter in Western Europe. Visiting crowed areas like the Netherlands result in conflicts between Bewick's swans and humans, because of for instance foraging on agricultural lands and the potential spread of avian influenza. In this project we investigated space use and foraging in winter by the Bewick's swans. We equipped 12 Bewick's swans with GPS-loggers to track then throughout the winter. We related their movements to both their infection status of a low-pathogenic avian influenza and to vegetation quality and growth.