MDI-GPU

MDI-GPU implements a GPU-accelerated Bayesian correlated clustering algorithm for integrated clustering of multiple large-scale and genomic-scale datasets.

Key Features:

• Bayesian correlated clustering: Enhanced implementation of a Bayesian correlated clustering algorithm for integration across multiple datasets.
• GPU acceleration: Leverages GPU-based computation to improve runtime performance by nearly four orders of magnitude.
• Scalability: Designed to handle datasets containing tens of thousands of items and genomic-scale analyses.
• Flexible Bayesian framework: Allows flexible modeling without relying on stringent assumptions.
• Integrated clustering capability: Facilitates correlated clustering across diverse, multi-dimensional datasets.

Scientific Applications:

• Integrated clustering: Perform integrated clustering across diverse datasets to uncover patterns and relationships in large-scale genomic data.
• Systems biology: Support systems biology analyses that require holistic integration across multiple data types.
• Genomic medicine: Enable analysis of genomic-scale datasets relevant to genomic medicine research.

Methodology:

GPU-accelerated implementation of a Bayesian correlated clustering algorithm using a flexible Bayesian modeling framework.