I found two articles interesting to me in today's issue of PLoS Computational Biology - the first one about becoming a good scientist, the other on circadian rhythms:
In the vein of promoting further debate and discussion, I provide here a different and perhaps deeper look at what makes a successful scientist. While I can't claim to have the reputation of Hamming, I grew up in a family of well-known scientists, and have had plenty of chances to observe the trajectories of scientific careers over my lifetime. Based on that experience, I propose the following as a somewhat distinct set of guidelines for doing the best research:
Life on earth is subject to daily light/dark and temperature cycles that reflect the earth rotation about its own axis. Under such conditions, organisms ranging from bacteria to human have evolved molecularly geared circadian clocks that resonate with the environmental cycles. These clocks serve as internal timing devices to coordinate physiological and behavioral processes as diverse as detoxification, activity and rest cycles, or blood pressure. In insects and vertebrates, the clock circuitry uses interlocked negative feedback loops which are implemented by transcription factors, among which the heterodimeric activators CLOCK and CYCLE play a key role. The specific DNA elements recognized by this factor are known to involve E-box motifs, but the low information content of this sequence makes it a poor predictor of the targets of CLOCK/CYCLE on a genome-wide scale. Here, we use comparative genomics to build a more specific model for a CLOCK-controlled cis-element that extends the canonical E-boxes to a more complex dimeric element. We use functional data from Drosophila and mouse circadian experiments to test the validity and assess the performance of the model. Finally, we provide a phylogenetic analysis of the cis-elements across insect and vertebrates that emphasizes the ancient link between CLOCK/CYCLE and the modeled enhancer. These results indicate that comparative genomics provides powerful means to decipher the complexity of the circadian cis-regulatory code.