The robust timing, or phase, of the circadian clock is critical in directing and synchronizing molecular, cellular, and organismal behaviors. The clock’s failure to maintain precision and adaption is associated with sleeping disorders, depression, and cancer. To better study and control the timing of circadian rhythms, we make use of systems theoretic tools such as sensitivity analysis and model predictive control (MPC). Sensitivity analysis is used to identify key driving mechanisms without having to fully understand or investigate the detailed mechanistic interconnections of the large complex circadian network. Contrary to intuition, sensitivity analysis of the circadian model highlights several non-photic control inputs (such as transcriptional regulation) that outperform light-based circadian phase resetting – light is known to accelerate protein degradation. Aside from targeting individual parameters as control inputs, our methods identify combinations of control targets that may further the efficiency of entrainment. We compare the phase resetting performance of our MPC algorithm among cases involving individual and multiple simultaneous control targets (in wild-type simulations). We then tailor the algorithm to correct continuously the phase mismatch that occurs in short and long period mutant phenotypes. Through use of the presented tools, our algorithm is robust in the presence of model mismatch and outperforms the natural in silico sun-cycle-based phase recovery strategy by nearly 3-fold.
Intuitive notions of brain-behavior relationships would suggest that because children show more variability in behavior, their brains should also be more variable. We demonstrate that this is not the case. In measuring brain signal variability with EEG and behavior in a simple face recognition task, we found that brain signal variability increases in children from 8-15 y and is even higher in young adults. Importantly, we show that this increased brain variability correlates with reduced behavioral variability and more accurate performance. A brain that has more variability also has greater complexity and a greater capacity for information processing. The implication of our findings is that variability in brain signals, or what some would call noise, is actually a critical feature of brain function. For the brain to operate at an optimal level, a certain amount of internal noise is necessary. In a certain way it could be stated that a noisy brain is a healthy brain.
The high-throughput sequencing of messenger RNA from parasitic organisms has permitted large-scale sequence analyses typically reserved for complete genome studies. Such expressed sequence tags (ESTs) have previously been generated for 37 species from the phylum Nematoda, of which 35 were from parasitic species. These datasets were combined with the complete genomes of Caenorhabditis elegans and C. briggsae. The sequences were assembled into 65,000 protein families, and decorated with 40,000 distinct protein domains. These annotations were analysed in the context of the nematode phylogeny. We identified massive gene loss in the model nematode, C. elegans, as well as plant-like proteins in nematodes that cause crop damage. Furthermore, many protein families were found in small groups of closely related species and may represent innovations necessary to sustain their parasitic ecologies. All of these data are presented at NemBase (www.nematodes.org) and will aid researchers working on this important group of parasites.
In microarray data analysis, when statistical testing is applied to each gene individually, one is often left with too many significant genes that are difficult to interpret or too few genes after a multiple comparison adjustment. Gene-class, or pathway-level testing, integrates gene annotation data such as Gene Ontology and tests for coordinated changes at the system level. These approaches can both increase power for detecting differential expression and allow for better understanding of the underlying biological processes associated with variations in outcome. We propose an alternative pathway analysis method based on mixed models, and show this method provides useful inferences beyond those available in currently popular methods, with improved power and the ability to handle complex experimental designs.