Post provided by Michael Morrissey

©Dr. Jane Ogilvie, Rocky Mountain Biological Laboratory

Evolutionary quantitative genetics provides formal theoretical frameworks for quantitatively linking natural selection, genetic variation, and the rate and direction of adaptive evolution. This strong theoretical foundation has been key to guiding empirical work for a long time. For example, rather than generally understanding selection to be merely an association of traits and fitness in some general way, theory tells us that specific quantities, such as the change in mean phenotype within generations (the selection differential; Lush 1937), or the partial regressions of relative fitness on traits (direct selection gradients; Lande 1979, Lande and Arnold 1983) will relate to genetic variation and evolution in specific, informative ways.

These specific examples highlight the importance of the theoretical foundation of evolutionary quantitative genetics for informing the study of natural selection. However, this foundation also supports the study other critical (quantification of genetic variation and evolution) and complimentary (e.g., interpretation when environments, change, the role of plasticity and genetic variation in plasticity) aspects of understanding the nuts and bolts of evolutionary change.

©Jia-Hong Chen

Evolutionary quantitative genetic theory is by no means set. Henshaw and Zemel (2017) draw on a range of new ways of looking at how natural selection changes the distribution of traits. They relate some of these to known quantities, thus foraging some key links to existing theory on the nuts and bolts of evolution. Chevin (2015) examines the ability of purely phenotypic notions of inheritance to generalise what is known about the dynamics of polygenic traits and the role of these dynamics in adaptive evolution.

One of the ongoing challenges in evolutionary quantitative genetics is to understand the best ways to use statistics as the link to connect theory to the realities of field data, and conversely, to understand the limitations of field data for informing theory. Adams (2016) address the issue of characterising modularity: if blocks of traits are closely (genetically) correlated with one another, then their short- to medium-term evolution is highly intertwined. This paper does some of the important core work in a constantly developing field of critically evaluating existing measures and providing new ones to fill gaps as they are identified. Similarly, Grabowski and Porto (2017) look at the ability of typical sample sizes used studies to characterise (phenotypic) measures of the multivariate geometry of trait covariation. They investigate the performance of particularly elegant measures of multivariate trait variation, that are well-grounded in evolutionary theory (Hansen and Houle, 2008).

© Vasco Elbrecht

At least for phenotypic measures of multivariate trait variation, these measures behave well in practice: great! Studies investigating the practical performance of these measures when applied to genetic variance-covariance matrices might be very useful too. In similar veins, Marrot et al. (2015) investigate the influence of spatial autocorrelation in fitness on measures of selection, and investigate ways to make analyses more robust; Waller and Svensson (2016) provide a similar treatment the effects and alleviation of imperfect detection on inference of viability selection; and Anthes et al. (2016) provide an important survey of issues surrounding measurement of sexual selection. Some challenges of course are more technical in nature than analytical, and  important work by McGoey et al. (2007) provides important progress in conducting controlled matings in wind-pollinated plants.

An important feature of the theoretical nuts and bolts of evolution provided by quantitative genetics is not just that it supports studies of genetic variation and natural selection, but provides links to other aspects of ecology and evolution. Khabbazian et al. (2016) present a step forward for linking models that have known relations to generation to generation evolutionary processes (see for e.g., Chevin and Haller 2014; Chevin et al. 2015) to patterns of phylogenetic diversification. Landguth et al.’s (2017) simulation platform could prove very useful for helping to incorporate evolutionary mechanisms with other meta-population processes.

This is an area where empirically-, statistically-, and theoretically-minded people could probably continue to come together, and Methods in Ecology and Evolution‘s strong developing tradition of publishing their works is very encouraging for future progress in evolution evolutionary quantitative genetics and at its intersection with other aspects of ecology and evolution.

All articles in the Evolutionary Quantitative Genetics Virtual Issue are freely available for a limited time.