birte

birte infers the influence of regulatory elements such as transcription factors (TFs) and microRNAs (miRNAs) on gene expression and reconstructs regulatory networks using a Bayesian modeling framework.

Key Features:

• Bayesian Inference Framework: Uses Bayesian inference to model regulator activities and provide probabilistic estimates that capture uncertainty in the data.
• Markov-Chain-Monte-Carlo (MCMC) Inference: Applies Markov-Chain-Monte-Carlo methods for parameter estimation and posterior exploration.
• Integration with Omics Data: Integrates mRNA and miRNA expression data into the probabilistic model to improve inference of regulatory influences.
• Network Reverse Engineering: Reconstructs potential regulatory networks to identify active regulators and their target genes from inferred activities.
• Simulation Studies and Real-World Applications: Validated through simulation studies and applied to contexts including prostate cancer and Escherichia coli growth control.
• Agreement with Biological Literature: Predicted regulatory networks generally concord with reported interactions in the biological literature.

Scientific Applications:

• Cancer Research: Infers regulator influences in cancerous tissues, exemplified by applications in prostate cancer research.
• Microbial Growth Studies: Elucidates regulatory networks controlling growth in microbial systems such as Escherichia coli.

Methodology:

Implements a Bayesian modeling framework and performs inference using Markov-Chain-Monte-Carlo (MCMC) methods, integrating mRNA and miRNA expression data to infer regulator activities and reverse-engineer regulatory networks.