Here is part 3 of my report on the 2011 BSCB-BSDB Spring Conference this April in Canterbury. In the first part, I covered Mark Krasnow’s amazing opening lecture on lung development, and in part two I introduced this year’s awardees of the BSCB and BSDB honorary medals.
Here I’ll highlight some of the talks in which the researchers used mathematical or computational modelling to understand and predict the behaviour of their system – in modelling terms they “explored the model’s parameter space”. Modelling your favourite biological system is very popular these days, which might be one of the reasons why a well-attended lunchtime workshop on the topic took place at the meeting. Its main take-home message was that modelling usually takes a lot longer than one might imagine, and therefore one shouldn’t underestimate the need for stable long-term collaborations or having a modeller or programmer in the lab.
The talks involving modelling spanned an array of topics, an indication of the widespread implications of the discipline. The subjects ranged from Enrico Coen‘s (John Innes Center, Norwich, UK) beautiful analyses of growth rates and the establishment of polarity during Arabidopsis leaf development, through Yogi Jaeger‘s (CRG, Barcelona, Spain) and Thomas Gregor‘s (Princeton University, USA) models of fly embryonic development, to Kees Weijer‘s (University of Dundee, UK) computational models of chemotactic cell migration in Dictostylium development and chick gastrulation. Here I’ll focus on just two of the studies.
Marie-Anne Félix (Institut Jacques Monod, CNRS-Université Paris Diderot, Paris, France) presented her lab’s recent work on the effect of quantitative variation within the intercellular signalling network that underlies C. elegans vulval development. Using a computational model of this network they varied the parameters of the model, without altering the network’s architecture, and analysed the phenotypic outcomes. A large fraction of the solutions turned out to result in the wild type pattern of vulva cell specification. Previously, two competing models of vulva cell precursor induction – one morphogen-like, the other involving a sequence of direct cell-cell signalling events had been proposed. Examining the parameter sets generated by the computational analysis revealed that they corresponded to either one or the other, or a mixture of the two mechanisms. This suggested that the two experimentally proposed mechanisms can function independently or in concert via the same network topology with the relative contribution of the two mechanisms depending on the quantitative tuning of parameters. Finally, Marie-Anne showed that inter-species differences in vulval patterning can be achieved by quantitative modulations of the very same network.
Denis Headon‘s lab (The Roslin Institute, University of Edinburgh, UK) is interested in vertebrate skin field patterning: Regional differences in the skin’s periodic micropattern of hair or feather follicles constitute the skin’s macropattern. To tackle the molecular pathways underlying macropattern generation, they took advantage of the Naked neck mutant, a naturally occurring chick mutant with a bare neck and lower overall feather density. Having pinned down the insertion associated with the trait, they identified increased levels of BMP12/GDF7 as the cause of the naked neck phenotype. They found that neck skin is more sensitive to BMP than the rest of the body since it selectively produces retinoic acid, which amplifies the effect of elevated BMP. This results in the complete loss of neck feathers. To validate the hypothesis, Denis showed the results of a mathematical model of the reaction-diffusion system of inhibitors and activators known to produce the periodic micropattern of feathers. Testing whether varying inhibitor (BMP) sensitivities, might lead to the macropatterning phenomenon observed in the mutant confirmed that a sharp pattern boundary between neck and body could explain the differences in feather density between neck and body, both in the wild type and at different BMP signalling levels. The simulated and experimental patterns astonished me as they looked more like graphic design than biology – it’s definitely worth having a look!
With these talks, the meeting provided examples of how modern developmental biology is indeed alive and kicking. To include modelling in the analysis is clearly one of the effective directions the field is taking, and in my final post on the Canterbury meeting I’ll highlight one of the other fruitful directions – live imaging.
Hoyos E, Kim K, Milloz J, Barkoulas M, Pénigault JB, Munro E, & Félix MA (2011). Quantitative variation in autocrine signaling and pathway crosstalk in the Caenorhabditis vulval network. Current biology : CB, 21 (7), 527-38 PMID: 21458263
Mou C, Pitel F, Gourichon D, Vignoles F, Tzika A, Tato P, Yu L, Burt DW, Bed’hom B, Tixier-Boichard M, Painter KJ, & Headon DJ (2011). Cryptic patterning of avian skin confers a developmental facility for loss of neck feathering. PLoS biology, 9 (3) PMID: 21423653