BUSseq

BUSseq fits an interpretable Bayesian hierarchical model to correct batch effects in single-cell RNA sequencing data in the presence of unknown cell types.

Key Features:

• Batch Effect Correction: Simultaneously corrects batch effects while accounting for unknown cell types.
• Cell Type Clustering: Clusters cell types as part of the integrated modeling framework.
• Count Data Modeling: Handles the count data nature of scRNA-seq data.
• Overdispersion Handling: Accounts for overdispersion in scRNA-seq data.
• Dropout Event Modeling: Incorporates dropout events characteristic of scRNA-seq data.
• Cell-Specific Sequencing Depth: Adjusts for cell-specific sequencing depth variation.
• Cross-Platform Integration: Integrates read count matrices obtained from different scRNA-seq platforms.
• Incomplete Cell Type Measurement: Allows cell types to be measured in some but not all batches under appropriate experimental design conditions.

Scientific Applications:

• Single-Cell RNA Sequencing Analysis: Enables downstream analysis of batch-corrected scRNA-seq data as if all cells were sequenced in a single batch.
• Multi-Batch Data Integration: Facilitates integration of scRNA-seq data across multiple batches or sequencing platforms.

Methodology:

BUSseq implements an interpretable Bayesian hierarchical model, the Batch Effects Correction with Unknown Subtypes for scRNA seq Data (BUSseq) model.