The Automatic Acquisition, Evolution and Reuse of Modules in Cartesian Genetic Programming

This paper presents a generalization of the graph- based genetic programming (GP) technique known as Cartesian genetic programming (CGP). We have extended CGP by utilizing automatic module acquisition, evolution, and reuse. To benchmark the new technique, we have tested it on: various digital circuit problems, two symbolic regression problems, the lawnmower problem, and the hierarchical if-and-only-if problem. The results show the new modular method evolves solutions quicker than the original nonmodular method, and the speedup is more pronounced on larger problems. Also, the new modular method performs favorably when compared with other GP methods. Analysis of the evolved modules shows they often produce recognizable functions. Prospects for further improvements to the method are discussed.     

Development of a generic activities model of command and control

This paper reports on five different models of command and control. Four different models are reviewed: a process model, a contextual control model, a decision ladder model and a functional model. Further to this, command and control activities are analysed in three distinct domains: armed forces, emergency services and civilian services. From this analysis, taxonomies of command and control activities are developed that give rise to an activities model of command and control. This model will be used to guide further research into technological support of command and control activities.     

Human pathogenic streptococcal proteomics and vaccine development

Gram-positive streptococci are non-motile, chain-forming bacteria commonly found in the normal oral and bowel flora of warm-blooded animals. Over the past decade, a proteomic approach combining 2-DE and MS has been used to systematically map the cellular, surface-associated and secreted proteins of human pathogenic streptococcal species. The public availability of complete streptococcal genomic sequences and the amalgamation of proteomic, genomic and bioinformatic technologies have recently facilitated the identification of novel streptococcal vaccine candidate antigens and therapeutic agents. The objective of this review is to examine the constituents of the streptococcal cell wall and secreted proteome, the mechanisms of transport of surface and secreted proteins, and describe the current methodologies employed for the identification of novel surface-displayed proteins and potential vaccine antigens.     

Pitfalls in quantitative contrast echocardiography: the steps to quantitation of perfusion

Current methods used clinically to assess myocardial perfusion are invasive and expensive. As the technology of ultrasound imaging improves, CE may provide a relatively inexpensive, noninvasive means of quantitating myocardial perfusion. Issues regarding stability of microbubble contrast agents must be studied more closely under physiologic conditions. As such, encapsulated microbubbles may provide more stability under physiologic pressures than free gas microbubbles. Introducing high concentrations of contrast, either by hyperconcentrating the contrast agent or by increasing the injection rate, may provide greater stability under physiologic conditions. Further, before quantitative statement of tissue perfusion can be made, the relationship between tracer concentration and system response must be established. Further, a "linear" postprocessing ultrasound setting does not eliminate this requirement as data must still undergo nonlinear transformation during log compression and time-gain compensation. Additionally, issues regarding "electronic thresholding" must be explored more extensively in vivo. Commercial ultrasound scanners, in their present form, may not offer adequate sensitivity for absolute quantitative studies. Further development of modified ultrasound systems may provide sufficient sensitivity for quantitative perfusion imaging. CE offers a potentially powerful tool in the clinical management of patients with ischemic heart disease. Conventional coronary angiography provides information on the size of a lesion, but accompanying tissue perfusion distal to the lesion cannot be determined. Doppler ultrasonography determines velocity of blood flow in large vessels but does not offer the potential to quantitate tissue perfusion. Clearly, CE has a place in the future of diagnostic imaging. The recent work of Ito et al. demonstrated the qualitative potential of CE in the identification of "areas at risk" in patients who had undergone thrombolysis or percutaneous transluminal coronary angioplasty after an acute myocardial infarction. With further improvement in the ultrasound imaging techniques and microbubble stability, CE may offer an inexpensive, noninvasive means of assessing myocardial perfusion.     

Activation of the mouse heart adenosine 5',5"'-P1-P4-tetraphosphate receptor

We have previously demonstrated that mouse brain membrane fractions have a specific, saturable receptor for diadenylated nucleotides. Binding is specific for two adenosines, and the length of the phosphate bridge is critical, with four phosphates being optimal [Hilderman et al. (1991) J. Biol. Chem. 266, 6915-6918]. In this report, we demonstrate that adenosine 5',5"'-P1,P4-tetraphosphate (Ap4A) binding to its receptor is dependent upon an activation step that requires divalent cations and a serine protease. Monoclonal antibodies (Mabs) are identified that inhibit Ap4A binding to its membrane receptor. These antibodies recognize a 212-kDa membrane protein. However, SDS-PAGE analysis of Ap4A cross-linked to membrane fractions reveals that Ap4A is not attached to the 212-kDa peptide but to a 30-kDa polypeptide. Appearance of the 30-kDa polypeptide is dependent on the activation step, and one of the inhibitory antibodies blocks its appearance. We suggest that the protease-dependent processing step involves cleavage of the 212-kDa component with the appearance of an active 30-kDa receptor.     

Nature of the porosity in American coals

The nature of the porosity in a number of 40 × 70 (i.e. − 40 + 70) mesh size American coals, varying in rank from anthracite to lignite, has been studied using the following characterization techniques: gas adsorption, helium and mercury displacement, and mercury porosimetry. Surface areas calculated from carbon dioxide adsorption at 298 K are consistently higher than those calculated from nitrogen at 77 K, indicating the molecular sieve character of the coals. Total pore volumes have been measured in the diameter range 12–29 600 Å and then divided into macropores (300-29 600 Å), transitional pores (12–300 Å), and micropores (4–12 Å). Coals vary widely in their total pore volumes (porosities range between 4.1 and 23.2%). In the lower-rank coals (carbon content less than 75%), porosity is primarily due to the presence of macropores. In coals having a carbon content in the range 76–84%, about 80% of the total open pore volume is due to micro and transitional pores, whereas in the coals of higher carbon content microporosity predominates. Coals having about 35–55% of their total open pore volume in the transitional range are expected to be most suitable for use for adsorption of organic molecules from solution. Thus, they are of interest as possible materials to be used in water purification. In general, these results are considered to be of importance in understanding the extent and rate of interaction of coals with gases and liquids.