ADVANCED STOCHASTIC MODELS FOR DNA MICROSATELITE SOMATIC INSTABILITY

这个子项目是许多研究子项目中利用 资源由NIH/NCRR资助的中心拨款提供。子项目和 调查员(PI)可能从NIH的另一个来源获得了主要资金， 并因此可以在其他清晰的条目中表示。列出的机构是 该中心不一定是调查人员的机构。 在人类和其他灵长类动物中，正常衰老会导致DNA损伤的积累，DNA的进行性退化，以及神经元的功能损伤和死亡。最近的全基因组研究发现，正常的人类衰老包括转录在核DNA修复途径中起关键作用的基因的下调。错误的DNA修复与高度可变的扩展重复序列的不稳定性有关，导致了各种遗传传播的人类神经疾病。 这项研究推进了一个正在进行的项目，开发和实施随机数学模型，分析现有的等位基因频率数据，测量体细胞组织中致病微卫星的进行性突变不稳定性。我们当前一代的连续时间马尔可夫链模型通过检查单一疾病(Friedreich共济失调)的单个大型数据集，揭示了以前未知的突变过程的特征。这一成功使我们获得了适合于类似建模分析的其他疾病(亨廷顿氏病和SCA7)的数据集。建模过程需要复杂且昂贵的非线性优化，包括重复评估模型拟合，因为参数在从4到10维的大的、混合的连续-离散的可行参数空间上变化。 将探索三种不同的方法以降低成本促进分析：开发新的混合优化算法，将“自然”(例如，遗传和+群)算法与传统优化方法结合使用，使用蒙特卡罗模拟，以及高吞吐量(并行)计算。开发的算法将针对几个现有的数据集进行测试，这两种方法都对DNA修复过程的年龄相关性提出了新的见解，并提供了新的工具来研究DNA微卫星体细胞不稳定的修复机制。

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Normative aging in humans and other primates leads to the accumulation of DNA lesions, progressive degeneration of DNA, and the functional impairment and death of neurons. Recent whole genome studies have found that normative human senescence includes down-regulation of genes whose transcripts play critical roles in nuclear DNA repair pathways. Erroneous DNA repair has been linked to the instability of the highly variable expanded repeat sequences causing a variety of genetically transmitted human neurological diseases. This study advances an ongoing project developing and implementing stochastic mathematical models that analyze existing allele frequency data measuring progressive mutational instability of pathogenic microsatellites in somatic tissues. Our current generation of continuous-time Markov chain models has revealed previously unknown characteristics of the mutation process by examining a single large data set for a single disease (Friedreich ataxia). This success has lead us to obtain data sets for other diseases (Huntington's and SCA 7) suitable for similar modeling analysis. The modeling process requires a complex and expensive nonlinear optimization involving repeated evaluation of model fit as parameters vary over a large, mixed continuous-discrete, feasible parameter space of from 4 to 10 dimensions. Three distinct approaches will be explored to facilitate the analysis at reduced expense: development of novel hybrid optimization algorithms that employ "natural" (e.g. "genetic' and +swarm algorithms in conjunction with traditional optimization methods, use of Monte-Carlo simulations, and high throughput (parallel) computation. The algorithms developed will be tested against several existing data sets, both suggesting new insights into the age dependence of the DNA repair process and providing new tools with which to study the repair mechanisms underlying DNA microsatellite somatic instability.