lemon-tree

lemon-tree infers module networks and upstream regulatory programs from integrated multi-omics datasets (gene expression, copy number variation, microRNA profiles, and epigenetic markers) to reconstruct co-expression modules and identify regulatory drivers.

Key Features:

• Integration of heterogeneous data types: Integrates gene expression, copy number variation, microRNA profiles, and epigenetic markers for joint analysis.
• Ensemble methods for inference: Uses ensemble-based algorithms to enhance the robustness and accuracy of module network inference.
• Reconstruction of co-expression modules and upstream regulators: Reconstructs co-expression modules and identifies upstream regulatory programs that govern module activity.
• Benchmarking on tumor datasets: Validated through benchmarking on large-scale tumor datasets, including analyses of The Cancer Genome Atlas (TCGA).
• Identification and prediction of driver genes: Recovers known oncogenes and tumor suppressors in glioblastoma samples and predicts novel candidate driver genes for downstream tumor pathway and survival analyses.

Scientific Applications:

• Cancer genomics (TCGA analysis): Analysis of TCGA and other large tumor datasets to identify master regulators and regulatory modules underlying tumorigenesis.
• Glioblastoma regulatory analysis: Identification of known oncogenes and tumor suppressors within module networks in glioblastoma samples.
• Discovery of candidate driver genes: Prediction of novel candidate driver genes for subsequent tumor pathway and survival analyses.

Methodology:

Integration of gene expression, copy number variation, microRNA profiles, and epigenetic markers followed by ensemble-based module network inference to reconstruct co-expression modules and identify upstream regulatory programs.