blase

BLASE maps bulk RNA-seq data onto single-cell pseudotime trajectories using Spearman correlation with bootstrapping for confidence estimation.

Key Features:

• Pseudotime Discretisation: BLASE discretises pseudotime using several different methods and provides heuristics for hyperparameter selection.
• Correlation-Based Mapping: Uses Spearman correlation to infer the closest matching period of pseudotime to a bulk RNA-seq sample.
• Bootstrapped Confidence Intervals: Bootstrapping provides confidence intervals around the correlation to inform "strong" calls.
• Visualisation: Enables visualisation of mapping results.
• Developmental Correction: Identifies and corrects for developmental rate differences in differential expression analysis.

Scientific Applications:

• Bulk RNA-seq "Dating": Determines the pseudotime position of bulk RNA-seq samples relative to single-cell trajectories.
• scRNA-seq Cell Type Annotation: Annotates cell types in single-cell RNA-seq data using existing bulk RNA-seq data.
• Spatial Transcriptomics: Resolves spatio-temporal processes such as keratinocyte differentiation in 10x Visium spatial data.
• Developmental Phenotype Correction: Corrects for developmental phenotype differences in bulk RNA-seq experiments, as demonstrated in Plasmodium falciparum heat shock studies.
• Life Cycle Mapping: Maps time-course microarray data to pathogen life cycles, such as the 48-hour Plasmodium falciparum lifecycle.

Methodology:

BLASE discretises the pseudotime of a scRNA-seq trajectory, then uses Spearman correlation to infer the closest matching period of pseudotime to a bulk RNA-seq sample. Bootstrapping provides confidence intervals around the correlation.