Serial Expression Analysis (SEA) is a web site for the analysis of serial gene expression data. Serial data is understood as multifactorial experimental designs where one of the factors is a quantitative variable such as time or treatment dose. The site offers five different methodologies for the identification of genes and functional classes which significant changes across series.

The following table summarizes the main characteristics of the SEA algorithms:

NameStatistical StrategySelected Features Selection criterion
maSigProUnivariate RegressionGenes Genes with differential expression profiles
maSigFun Univariate Regression Functional CategoriesFunctional classes with most genes having correlated differential expression profiles
ASCA-genesANOVA + Multivariate ProjectionGenesGenes that follow major expression trends
ASCA-functionalANOVA + Multivariate Projection + GSAFunctional CategoriesFunctional classes associated to a given expression trend
PCA-maSigFunMultivariate Projection + Univariate RegressionFunctional Categories Functional classes with subset of genes showing correlated differential expression profiles

job listJob list
data listData list