Parametric Treatment of cDNA Microarray Data
Parametric Treatment of cDNA Microarray Data
复制标题
cDNA 微阵列数据的参数化处理
DOI：
--
复制
发表时间：
2002
期刊：
影响因子：
--
通讯作者：
T. Konishi
中科院分区：
文献类型：
--
作者：
T. Konishi
由于每组微阵列数据都受到实验条件变化的影响，因此需要适当的归一化处理。已经提出了各种实现这种规范化的方法，通常涉及对每对数据集的调整，通常在同时杂交的R和G探针之间。Chen等人（[1]）引入了一个模型，其中每个探头之间的信号比都是正态分布的；在斯坦福微阵列数据库[6]的过程中基本使用了相同的概念。最近，Yang等人（[4]）扩展了该方法，稳定了信号比的平均值，该平均值可以根据信号强度进行偏置。通过引入一个参数，这种非参数方法得到了进一步改进，该参数也保持了信号比的方差[2,3]。另一种假设数据为对数正态分布的简单参数方法也被广泛采用。除了易于执行计算之外，它还能够确定数据z分数，这可能是数据比较的常用单位。然而，微阵列数据往往具有偏态分布，这种低质量的分布模型严重限制了数据的准确性。在参数化过程中，背景估计被设计为加性噪声[1]的恒定部分，可能是归一化不准确的主要来源。在大多数情况下，从图像数据的DNA点外的区域估计背景；假设尖端的背景是均匀的。然而，由于DNA斑点和尖端的完整表面可以结合不同密度的游离染料，这种估计显然容易出错。为了检查这种可能性，一种基于对照DNA信号强度的替代估计方法与传统的基于图像的方法进行了测试。使用概率图调查了两组处理数据的分布。
As each set of microarray data is a ected by variations in experimental conditions, appropriate nor-malizationprocesses are required. Various approaches towards such normalization have been proposedand generally involve adjustments to every pair of the data sets, often between the simultaneouslyhybridizing R and G probes. Chen et al. [1] introduced a model in which each the signal ratio betweenthe probes is normally distributed; the same concept is basically used in the process of the StanfordMicroarray Database [6]. Recently, Yang et al. [4] extended this method by stabilizing the average ofthe signal ratio, which could be biased based on signal intensity. Such non-parametric methods havebeen further improved by introduction of a parameter that also maintains the variance in the signalratios [2, 3]. An alternative and simple parametric method that assumes lognormal distribution ofdata has also been widely employed. Besides its simple ease in performing the calculations, it is alsocapable of determining the data z-scores, a possible common unit for data comparisons. However, mi-croarray data often have a skewed distribution, and this low delity to the distribution model severelylimits the accuracy of the data.In the parametric process, the estimation of background, which is de ned as the constant part ofadditive noise [1], can be a major source of normalization inaccuracies. In most cases, the backgroundis estimated from the area outside the DNA spot of the image data; based on the assumption thatthe background on a tip is uniform. However, as DNA spots and also the intact surface of the tipcan bind free dyes at di erent densities, such estimations are clearly prone to errors. To check thispossibility, an alternative estimation method that is based on signal intensity of control DNA is testedagainst the conventional image-based method. The distributions of both sets of processed data areinvestigated using probability plots.