Topconfects

Topconfects computes biologically meaningful rankings of differential gene expression by estimating confidence bounds on log fold changes to prioritize genes with substantial expression changes.

Key Features:

• Ranking by confidence bounds: Ranks genes using confidence bounds around log fold changes rather than by p-value ordering.
• Error rate control: Controls false discovery rates and false coverage-statement rates for ranked effect sizes.
• TREAT-based inference: Derives confidence bounds from the TREAT test framework for differential expression.
• Implementation: Provided as an R package for integration into differential expression analysis workflows.

Scientific Applications:

• Cancer transcriptomics (e.g., breast cancer): Identifies top-ranked genes that emphasize biological processes different from p-value rankings, aiding pathway and mechanism discovery.

Methodology:

Confidence bounds on log fold changes are computed using the TREAT test framework with control of false discovery rates and false coverage-statement rates.