Modeling the evolution of transcriptional systems in silico
For the past few decades, the field of molecular evolution has largely focused on the evolution of individual gene families and overall genome structure. Analogous to the transition from reductionist to system-scale approaches in molecular biology, the logical next step in molecular evolution research is to study the evolution of genes in the context of the systems in which they function. One way to study how systems of interacting components evolve is to simulate the evolution of suitably abstracted system models in silico. Given recent developments in the field of high-performance computing, it is now possible to simulate the evolution of molecular systems at an unprecedented level of mechanistic detail. We use mathematical models to map the genotype of artificial transcriptional regulatory systems to their expression phenotype. We then use these models in population-scale evolutionary simulations to investigate how transcriptional systems evolve. One of our major research interests is to study how duplicated transcriptional systems diverge. We are particularly interested in unraveling the mechanisms by which dosage-balance constraints on duplicated transcription factors can be resolved over evolutionary time. Many regulatory genes do not duplicate easily on their own, because duplication upsets their dosage balance with other transcription factors or targets. Whole-genome duplication (WGD) on the other hand is thought to preserve dosage balance, and moreover to lead to selective retention of dosage-balance sensitive genes, since their loss after WGD would create a reverse dosage balance effect. As a consequence, post-WGD organisms are endowed with a 'regulatory spandrel’, a collection of regulatory genes that may not immediately add extra functionality but that cannot be purged easily from the genome. It is thought that under the right circumstances, these non-adaptively preserved genes may be co-opted for adaptive innovations, but this requires that the dosage-balance constraints be lifted, and it is currently not clear how this is accomplished. Profiling individual field-grown plants to reverse engineer plant systems and crosstalk between stress pathways
To develop crops with enhanced yield and tolerance to field stress conditions, we need to fundamentally change our approach to studying stress response pathways in plants. Most stress studies performed under controlled laboratory conditions are of limited predictive value for phenotypes in the field. In lab conditions, stress responses are usually studied in isolation, whereas in the open environment a multitude of stresses operate in synergistic and antagonistic interaction to modulate plant phenotypes. To get a view on the complex interactions between plant stress response pathways and their effect on yield phenotypes, we aim to harness natural gene expression and phenotype variation among genetically identical field-grown plants, based on the premise that each individual plant is subject to subtle deviations of several micro-environmental factors from the field average. We are currently developing methods to use individual plant datasets for reverse engineering stress response pathways, their interaction and their impact on yield-related phenotypes, focusing on maize and rapeseed as model crops.