STATISTICAL MODELS IN TOXICOLOGY AND BIOCHEMISTRY

该项目旨在增加我们对使用的理解， 模型在毒理学和生物化学中的应用， 数学模型，以帮助解释目前的研究结果。的 研究工作探索了各种生物领域，包括 致癌作用、药理学、发育生物学和免疫学。在 致癌作用，致癌作用的多阶段模型的适当性， 评估未处理动物中肿瘤发病率的估计。还有， 根据动物实验对致癌能力的估计， 化学相关的生存变化已经开发出来， 评估。一个致癌作用的数学模型， 将DNA修复纳入多阶段过程的技术正在开发中。 数学细节使我们能够使用该模型已经开发 目前的研究集中在使用数据来实现这个模型 例如化学品。在另一个研究领域， 实验，克隆两阶段模型的影响， 对两年啮齿动物生物测定设计进行了评价。一个简单 使用年龄依赖性给药方案的设计产生了独特的肿瘤模式 发病率，可以提供更好的致癌机制的信息。 这项研究正在继续进行，重点是在阶段的数量， 致癌过程我们也在研究致癌的形状 剂量-反应曲线使用的生物活性的信息， 化工有限选择适当的形状（阈值或不）对于 准确估计致癌风险。在畸形学中， 评估模型与现有数据的一致性进行了评价。没有模型 被认为是足够的，有些被发现产生误导性的结果。 畸形学的另一项努力涉及恢复技术的能力， 和准似然方法来解释窝间相关性。 对死亡率和发病率风险估计的进一步研究 免疫抑制化学物质的影响已经开始了。

This project is intended to increase our understanding of the use and application of models in toxicology and biochemistry and to implement new mathematical models to aid in explaining current research findings. The research effort explores a diverse range of biological areas including carcinogenesis, pharmacology, developmental biology and immunology. In carcinogenesis, the adequacy of the multistage model of carcinogenesis for estimating tumor incidence rates in untreated animals was evaluated. Also, an estimate of carcinogenic potency from animal experiments which adjusts for chemically related changes in survival has been developed and is being evaluated. A mathematical model of carcinogenesis which explicitly incorporates DNA repair into the multistage process is under development. The mathematical details allowing us to use the model have been developed and current research is focused on using data for implementing this model for example chemicals. In another research area, the design of animal experiments, the implications of a clonal two-stage model of carcinogenesis on the design of the two-year rodent bioassay was evaluated. A simple design using age-dependent dosing schemes produced unique patterns of tumor incidence which can provide better information on carcinogenic mechanism. This research is continuing with emphasis on the number of stages in the carcinogenesis process. We are also studying the shape of carcinogenesis dose-response curves using information on biological activity of the chemical . Choosing the proper shape (threshold or not) is critical to an accurate estimate of carcinogenic risk. In teratology, several risk assessment models were evaluated for agreement with existing data. No model was found to be adequate and some were found to produce misleading results. Another effort in teratology concerns the ability of resampling techniques and quasi-likelihood methods to account for interlitter correlations. Additional research into the estimation of risks of mortality and morbidity from exposure to immunosuppressive chemicals has begun.