Cinquin Lab

Research philosophy

The main characteristic that a scientist studying natural phenomena must have is complete freedom of the spirit based on philosophical doubt […] . When we conceive a general scientific theory, the one thing of which we can be certain is that — speaking in absolute terms — all such theories are false. They are but partial and provisional truths that we need, like steps on which we rest, to advance in our investigation. […] If an idea arises, we should not reject it only because it does not agree with the logical consequences of a dominant theory. Claude Bernard (the first systems biologist), Introduction à l’Etude de la Médecine Expérimentale (Baillière, Paris, 1865). Translation source. More extensive excerpt in French.

One of the great successes of 20th-century biology has been the identification of genes and the exploration of their regulation. The discrete character of genes and powerful methods for reverse and forward genetics have provided a natural method to frame biological questions. It has become clear, however, that the collection of parts provided by this approach does not by itself provide a satisfactory understanding of system-level properties of cells, organs, organisms, and ecosystems. For example, despite the identification of many genes involved in the control of stem cells, in no model system (much less humans) is there a comprehensive understanding of the regulation of stem cell proliferation and differentiation, and of how that regulation is shaped by evolutionary forces. Such an understanding is important from both scientific and therapeutic standpoints, and will not automatically emerge once the last stem cell-related gene has been characterized.

We ask questions about stem cell proliferation and differentiation bearing in mind the following:
•    Cell differentiation is a dynamic process. Although this might seem to be a truism, only recently has it been possible to show in detail that the dynamics of gene expression matter for differentiation outcome. We investigate those dynamics experimentally and theoretically. In the C. elegans germ line, we are greatly helped by the arrangement of cells in a presumptive gradient of maturity, which makes it relatively straightforward to deduce temporal evolution from spatial expression patterns. Temperature-sensitive mutations and the use of drugs also make it possible to follow system response to perturbations applied with great temporal precision, which is more informative than looking at a broken system at steady state.
•    Regulatory networks are horrendously more complex than one would naïvely expect. Using our crude lab assays, it is often necessary to use double or triple mutants to reveal functions of a given gene. One obvious cause of this “redundancy” is that we are missing important system-level behaviors that are important to animal fitness in the wild. As Lewis Wolpert puts it, “In mice, there are many genes that you knock out and you don’t see a phenotype, and one concludes that they are redundant. I say, have you taken your mice to the opera? Can they still tell Wagner from Mozart?”. Understanding complex regulatory networks is probably hopeless if we don’t know what functions they are fulfilling. Defining those functions is no easy task because fitness of mutants is difficult to observe in the wild rather than under laboratory conditions. Experimental evolution and culture in conditions more representative of the wild have emerged as useful proxies, which can be used to study the C. elegans germ line.
•    Even with the development of high-throughput data acquisition methods, there will never be enough experimental data. A key problem is guiding data acquisition in a meaningful way. As a concrete example, the set of known interactions between elements of regulatory networks is likely incomplete, even in well-characterized networks. The minute characteristics of those interactions are also often not known. It is crucial to be able to predict which potential or established interactions are more important to interesting behaviors of the system, in order to focus experimental attention.