Methods for evaluating gene expression from Affymetrix microarray datasets.

摘要: Background: Affymetrix high density oligonucleotide expression arrays are widely used across all fields of biological research for measuring genome-wide gene expression. An important step in processing microarray data is to produce a single value the level an RNA transcript using one growing number statistical methods. The challenge researcher decide on most appropriate method use address specific question with given dataset. Although several efforts have focused assessing performance few methods evaluating from hybridization experiments different datasets, relative merits currently available literature collected real remain actively debated. Results: present study reports comprehensive survey seven commonly well-designed experiment microarrays. profiled eight genetically divergent barley cultivars each three replicates. dataset so obtained confers balanced and idealized structure analysis. were evaluated their sensitivity detecting differentially expressed genes, reproducibility values replicates, consistency calling genes. genes detected as among differed by factor two or more at false discovery rate (FDR) level. Moreover, we propose containing feature polymorphisms (SFPs) empirical test comparison ability detect true differential basis that SFPs largely correspond cis-acting regulators. PDNN demonstrated superiority over other every comparison, whilst default MAS5.0 was clearly inferior. Conclusion: A assessment extraction based extensive has shown superior detection