BUScorrect

BUScorrect fits a Bayesian hierarchical model, the Batch-effects-correction-with-Unknown-Subtypes model (BUS), to correct batch effects in high-throughput experimental data in the presence of unknown subtypes.

Key Features:

• Batch Effect Correction: Explicitly corrects batch effects in high-throughput experimental data.
• Subtype Discovery: Groups samples that share similar characteristics into subtypes without prior knowledge.
• Feature Identification: Identifies features that distinguish discovered subtypes.
• Computational Efficiency: Achieves linear-order computation complexity for scalable analysis.

Scientific Applications:

• Public Database Mining: Enables valid scientific discovery from exponentially accumulating high-throughput experimental data in public databases by addressing technical artifacts and biological heterogeneity.

Methodology:

BUScorrect implements a Bayesian hierarchical model, the Batch-effects-correction-with-Unknown-Subtypes model (BUS).