Zhang Li, Miles Michael F, Aldape Kenneth D
Department of Biostatistics, The University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd. 447, Houston, Texas 77030, USA.
Nat Biotechnol. 2003 Jul;21(7):818-21. doi: 10.1038/nbt836. Epub 2003 Jun 8.
High-density short oligonucleotide microarrays have become a widely used tool for measuring gene expression on a large scale. However, details of the mechanism of binding on microarrays remain unclear. Short oligonucleotide probes currently synthesized on microarrays are often ineffective as a result of limited sequence specificity or low sensitivity. Here, we describe a model of binding interactions on microarrays that reveals how probe signals depend on probe sequences and why certain probes are ineffective. The model indicates that the amount of nonspecific binding can be estimated from a simple rule. Using this model, we have developed an improved measure of gene expression for use in data analysis.
高密度短寡核苷酸微阵列已成为一种广泛用于大规模测量基因表达的工具。然而，微阵列上结合机制的细节仍不清楚。目前在微阵列上合成的短寡核苷酸探针由于序列特异性有限或灵敏度低，往往效果不佳。在这里，我们描述了一种微阵列上结合相互作用的模型，该模型揭示了探针信号如何依赖于探针序列以及某些探针无效的原因。该模型表明，可以从一个简单规则估计非特异性结合的量。利用这个模型，我们开发了一种用于数据分析的改进的基因表达测量方法。
