Faculty Bibliography

This paper addresses questions regarding the decidability of hybrid automata that may be constructed hierarchically and in a modular way, as is the case in many exemplar systems, be it natural or engineered. Since an important step in such constructions is a product operation, which constructs a new product hybrid automaton by combining two simpler component hybrid automata, an essential property that would be desired is that the reachability property of the product hybrid automaton be decidable, provided that the component hybrid automata belong to a suitably restricted family of automata. Somewhat surprisingly, the product operation does not assure a closure of decidability for the reachability problem. Nonetheless, this paper establishes the decidability of the reachability condition over automata which are obtained by composing two semi-algebraic o-minimal systems. The class of semi-algebraic o-minimal automata is not even closed under composition, i.e., the product of two automata of this class is not necessarily a semi-algebraic o-minimal automaton. However, we can prove our decidability result combining the decidability of both semi-algebraic formulae over the reals and linear Diophantine equations. All the proofs of the results presented in this paper can be found in [1].

Current symbolic techniques for the automated reasoning over undecidable hybrid automata, force one to choose between the refinement of either an overapproximation or an underapproximation of the set of reachable states. When the analysis of branching time temporal properties is considered, the literature has developed a number of abstractions techniques based on the simulation preorder, that allow the preservation of only true universally quantified formulæ. This paper suggests a way to surmount these difficulties by defining a succession of abstractions of hybrid automata, which not only (1) allow the detection and the refinement of both over- and under-approximated reachable sets symmetrically, but also (2) preserves the full set of branching time temporal properties (when interpreted on a dense time domain). Moreover, our approach imposes on the corresponding set of abstractions a desirable monotonicity property with respect to the set of model-checked formulaæ.

Using the vital marker GFP and its spectral variants, it is possible to visualize multiple proteins in individual cells and thereby monitor embryonic development on a cellular and molecular level. In the following chapter we describe how to prepare Drosophila embryos or larvae for live imaging or immunohistochemical staining and provide some guidelines for optimal GFP detection.

The immunodominant epitope of bovine type II collagen (CII256-270) in Aq mice carries a hydroxylysine-264 linked galactose (Gal-Hyl264), the recognition of which is central to the development of collagen-induced arthritis. This study explores the molecular interactions involved in the engagement of T-cell receptors (TCRs) with such epitopes. Responses of three anti-CII T-cell hybridomas and clone A9.2 (all sharing close TCR sequences) to a panel of CII256-270 analogues incorporating Gal-Hyl264 with a modified side chain were determined. Recognition of naturally occurring CII256-270 peptides by either group of T cells depended strictly upon the presence of the carbohydrate and, more precisely, its intact HO-4 group. Modifications of primary amino group on the hydroxylysine side chain eliminated T-cell reactivity, notwithstanding the presence of the galactosyl moiety. Moderate stereochemical changes, such as altered sugar orientation and methylation at the galactose anchor position, were still permissive. Conversely, robust transformations affecting the relative positions of the key elements were detrimental to TCR recognition. To conclude, these data provide strong new experimental evidence that integrity of both galactose HO-4 and hydroxylysine side chain primary amino groups are mandatory for activation of anti-Gal-Hyl264 TCRs. They also indicate that there is a certain degree of TCR plasticity in peptide-TCR interactions.

INTRODUCTION: A gap exists between asthma guidelines and actual care delivered. We developed an educational intervention using simulated physician-patient encounters as part of a project to improve asthma management by community-based primary care providers. We hypothesized that this type of skills-based interactive training would improve learners' care choices for simulated patients after training compared with their choices before training. METHODS: After a pilot project was done on a small group of providers, a larger group of primary care providers (PCPs) was recruited to be trained with our interactive materials. The pilot session, with 39 providers, showed that the cases were felt to be appropriate, that the time allocated for discussion was adequate, that the models were useful, that the experience was educational, and that the experience captured their interest. Two subsequent training sessions were held with 240 PCPs. Participants completed a questionnaire to elicit perceived barriers and self-efficacy and then viewed a short simulated physician-patient dialogue. They then completed a set of scaled questions about treatment choices. This served as a pretest assessment. A similar simulation was then shown, and the group discussed their thoughts on diagnosis and treatment. Finally, they viewed another physician-patient interaction and responded to the same questions as posed for the pretest assessment; the responses before and after assessment were compared. RESULTS: Following completion of the intervention, providers were significantly (p < 0.05) more likely to make use of controller medications, asthma equipment, and patient training. Significant (p < 0.05) increases were also seen in action plan development and the availability of office visits. Providers were significantly (p < 0.05) less likely to refer asthma patients to an emergency department or for hospitalization. Significant (p < 0.05) improvements were also seen in perceptions of self-efficacy and barriers to treatment. There were significant (p < 0.05) increases in learners' confidence about their own and patients' abilities to improve asthma care, and fewer barriers to asthma management were reported after the training. DISCUSSION: This method of training resulted in learners showing a measurable improvement in their intent to follow guidelines as applied to simulated patients. An evaluation addressing actual patient outcomes will need to be done.