CAJ | 학술논문

如何分离出少量区别不同组织类型的特异性基因是DNA微阵列数据分析中的主要问题,特别是构建恰当的统计模型来刻画这些不同组织类型的DNA表达形式尤为重要.为此,基于基因DNA微阵列数据的特点,我们假定对数变换后的微阵列数据服从混合正态分布.我们采用分级Bayesian先验刻画不同基因的相关性,利用分级Bayesian方法构建模型,给出了刻画不同组织基因表达的差异的一个标准,用MCMC迭代计算该标准.模拟计算表明我们的模型具有较好的识别能力.
여하분리출소량구별불동조직류형적특이성기인시DNA미진렬수거분석중적주요문제,특별시구건흡당적통계모형래각화저사불동조직류형적DNA표체형식우위중요.위차,기우기인DNA미진렬수거적특점,아문가정대수변환후적미진렬수거복종혼합정태분포.아문채용분급Bayesian선험각화불동기인적상관성,이용분급Bayesian방법구건모형,급출료각화불동조직기인표체적차이적일개표준,용MCMC질대계산해표준.모의계산표명아문적모형구유교호적식별능력.
In DNA microarray analysis, there is often interest in isolating a few genes that best discriminate between tissue types. In particular, it is critical to develop suitable models to explain the patterns of DNA expression for these different types of tissues. Toward this goal, we propose a methodology for the analysis of high-density oligonucleotide arrays.The log-transformed data are assumed to follow a mixture normal distribution based on the characteristic of gene itself. The variation in the data can reasonably be thought to arise from the effects of genes, tissue types, and their interactions. We introduce a hierarchical Bayesian priors for the parameters and propose a model selection criterion for identifying subsets of genes that show different expression levels between normal and tumor types. In addition, we develop Markov chain Monte Carlo algorithms for sampling from the posterior distribution of parameters and for computing criterion. The proposed methodology is evaluated via simulations studies.