ASC

ASC applies an empirical Bayes approach to detect differential gene expression from high-throughput sequencing data by estimating posterior means of log fold change and associated false discovery rates.

Key Features:

• Empirical Bayes Framework: ASC pools information across sequences to establish a prior distribution that reflects biological variability.
• Accommodates limited replicates: The empirical Bayes approach enables accommodation of biological variability even in the absence of replicates.
• Reduced bias across abundance levels: The method mitigates bias toward highly expressed genes, enabling detection of differential expression across a broader range of gene abundances.
• Posterior log fold change estimation: ASC estimates the posterior mean of log fold change for each gene to represent both magnitude and statistical evidence of expression changes.
• False discovery rate estimation: ASC computes false discovery rates based on posterior means to control type I error in differential expression calls.

Scientific Applications:

• Transcriptome analysis: Detection of differential gene expression from RNA-seq and other high-throughput sequencing experiments.
• Studies with limited replicates: Analyses where biological variability must be estimated despite few or no replicates.
• Genomics, oncology, and developmental biology: Identification of gene regulation and functional changes relevant to genomic research, cancer studies, and developmental processes.

Methodology:

Inputs high-throughput sequencing count data; pools information across sequences to establish an empirical Bayes prior reflecting biological variation; computes posterior means of log fold change per gene integrating biological and technical variability; estimates false discovery rates from posterior means to identify differentially expressed genes.