scRecover

scRecover imputes dropout values in single-cell RNA-seq raw read counts matrices while preserving real zeros, distinguishing between technical dropout zeros and true biological zeros to improve downstream analysis accuracy.

Key Features:

• Dropout Detection: Distinguishes dropout zeros from real zeros in scRNA-seq raw read counts matrices.
• Imputation Integration: Combines scImpute, SAVER, and MAGIC to achieve higher accuracy in dropout and real zero detection.
• Downstream Analysis Improvement: Enhances clustering and visualization results following imputation.

Scientific Applications:

• Single-Cell RNA-Seq Data Imputation: Recovers gene expression values lost due to technical dropouts in scRNA-seq experiments.
• Cell Clustering Enhancement: Improves the accuracy of cell type identification and clustering in single-cell transcriptomic studies.
• Data Visualization: Generates improved visual representations of single-cell gene expression data after dropout correction.

Methodology:

scRecover detects and imputes dropout values in scRNA-seq raw read counts matrices by combining scImpute, SAVER, and MAGIC.