TCC

TCC performs differential expression analysis of RNA-seq count data using the DEGES (Differential Expression Gene Estimation Strategy) normalization approach to reduce bias by removing potential differentially expressed genes or transcripts prior to normalization.

Key Features:

• DEGES: Removes potential differentially expressed genes or transcripts from the dataset prior to normalization to prevent normalization bias.
• Robust normalization methods: Implements robust normalization strategies for count data through combinations of functions from dependent packages.
• Negative binomial modeling: Models count data using the negative binomial distribution for expression analysis.
• Dispersion estimation: Estimates dispersion parameters using a quantile-adjusted conditional maximum likelihood estimator (PMID: 17728317).
• Exact test: Derives an exact test for hypothesis testing that improves on approximate asymptotic tests.
• Small-sample applicability: Provides methods that perform well in very small sample sizes, such as serial analysis of gene expression studies.

Scientific Applications:

• Differential expression analysis: Identification of differentially expressed genes or transcripts from RNA-seq count data.
• Normalization of count data: Reducing normalization bias in RNA-seq and other count-based expression datasets by removing putative DEGs prior to normalization.
• Dispersion estimation in small experiments: Accurate dispersion estimation for negative binomial models in small-sample experiments.
• Statistical hypothesis testing: Conducting hypothesis tests for differential expression using an exact test framework.

Methodology:

DEGES removes putative DEGs prior to normalization; normalization uses combinations of functions from dependent packages; count data are modeled by the negative binomial distribution with dispersion estimated via a quantile-adjusted conditional maximum likelihood estimator (PMID: 17728317); an exact test for differential expression is derived.