Research

Baer Lab Research

I am a comparative evolutionary geneticist whose research is motivated by theoretical population genetics. My primary research interest is in the factors responsible for the generation and maintenance of genetic variation – “understanding variation in genetic variation”. I am especially interested in the evolution of mutation rate, and in the distribution of fitness effects of new mutations. There is considerable variation in the rate and cumulative effects of new mutations, even among genotypes within species. I begin from the premise that the rate and effects of mutation are themselves evolvable phenotypes which are subject to optimizing selection, and which may evolve in predictable ways. My research program has two primary objectives: (1) elucidate the various factors that underlie variation in the rate, molecular spectra and phenotypic effects of spontaneous mutations, and (2) determine the extent to which variation in mutational properties explains variation among taxa in standing genetic variation at the phenotypic and molecular level. Our studies of mutational variation have led me to become interested in the evolution of phenotypic robustness and epigenetic variation.

We use rhabditid nematodes as our experimental organism, and employ a variety of phenotypic and molecular methods to address the questions of interest. Additional research interests include the evolution of genetic architecture (i.e., genetic covariance), and selection experiments in any way, shape, or form. Recently, I have become involved in collaborative work with chemical engineers and cell biologists to employ experimental evolution to characterize the phenotypic and molecular effects of substrate rigidity on cultured mammalian cells.

Current projects in the lab include:

Is the mutation rate self-dependent?
Evidence is accumulating that the mutation rate increases under conditions of physiological stress. Deleterious mutations constitute an endogenous source of “stress”, which implies that the mutation rate may be an increasing function of itself, with interesting implications [1]. To test that hypothesis, we constructed sets of “second-order” mutation accumulation (MA) lines derived from “first-order” MA lines with high and low fitness. Whole genome sequencing combined with fitness assays reveal that the base-substitution rate is not fitness-dependent, but the indel rate is, but not in the expected way: high-fitness lines have higher (not lower) rates of indel mutations than low-fitness lines. The simplest explanation is that deleterious mutations interact synergistically. The total mutation rate increased in the second-order MA lines, which is consistent with the “drift barrier” hypothesis of the evolution of mutation rate [2]. Funded by NSF grant DEB-0717167 and NIH grant R01GM072639.

The relationship between standing genetic variance and mutational variance.
At mutation-selection balance, the standing genetic variation within a population for a trait (VG) is directly proportional to the mutational variance for the trait (VM) and inversely proportional to the strength of selection acting on the trait (S), or in other words, VM/VG S. Several studies, from our lab [3,4] and others, have shown that VM predicts VG for a variety of traits in worms and flies, but the quantitative relationship between VM/VG and S is uncertain. Recently, we combined high-throughput fitness assays with whole genome sequencing of MA lines to directly estimate the average strength of selection against spontaneous mutations [5]. Remarkably, the two methods agree to within twofold, from which we tentatively conclude that VM/VG may provide a quantitatively reliable estimate of the strength of selection acting on new mutations. We are pursuing this line of investigation for a variety of traits in C. elegans, and hope to extend the work to additional species. Funded by NIH grants R01GM072639 and R01GM107227.

Mutation, selection, phenotypic plasticity, and phenotypic noise
It has long been known that both mutations of large effect and severe environmental stress increase the non-genetic component of phenotypic variance (VE, or “noise”) relative to the wild-type in benign conditions, and we have shown that holds true for the cumulative effects of spontaneous mutations, for a variety of traits [6,7]. In other words, deleterious mutations “de-canalize” the phenotype. Intriguingly, gene expression is a possible exception: mutation accumulation appears to reduce the non-genetic variance in transcription [8].
It has been proposed that there is a fundamental tradeoff between the noise in an unperturbed system and the ability of the system to respond when perturbed (“plasticity”). Moreover, robustness (~ 1/”noise”) at an emergent level implies there must be plasticity at some underlying point of control. If the proposed noise/plasticity tradeoff is true, it suggests an explanation for the apparently anomalous relationship between mutation and noise in gene expression: natural selection favors robust phenotypes, which requires plasticity of the underlying controls, i.e., gene expression. Deleterious mutations reduce plasticity of gene expression and simultaneously increase noise in emergent phenotypes.
We are currently investigating the noise/plasticity relationship with respect to several measures of fitness in the context of variable thermal environments (i.e., hot and cold), and plans are afoot to rigorously test the hypothesis with respect to gene expression, in collaboration with Itai Yanai at NYU. Funded by NIH grant R01GM072639.

The Distribution of Fitness Effects (DFE) of spontaneous mutations
The DFE is a fundamental parameter in evolutionary biology, but its estimation presents formidable practical problems. In collaboration with Erik Andersen (Northwestern), Kim Gilbert (University of Bern) and José Miguel Ponciano at UF, we are combining high-throughput phenotyping and whole-genome sequencing with a classical quantitative genetic breeding design and novel statistical analyses to experimentally characterize the homozygous and heterozygous DFEs. Stay tuned… Funded by NIH grant R01GM107227.