PRIME: a probabilistic imputation method to reduce dropout effects in single-cell RNA sequencing.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Single-cell RNA sequencing (scRNA-seq) enables transcriptomic analysis at the individual cell level.
• Dropout effects, characterized by zero-inflated distributions, are a significant source of noise in scRNA-seq data, compromising reliability.
• Effective data processing is crucial for accurate downstream analysis of scRNA-seq profiles.

Purpose of the Study:

• To develop and evaluate a novel imputation method to mitigate dropout effects in scRNA-seq data.
• To enhance the reliability and interpretability of scRNA-seq datasets.
• To improve the accuracy of visualization and clustering analyses.

Main Methods:

• A probabilistic imputation method, PRIME (PRobabilistic IMputation to reduce dropout effects in Expression profiles of single-cell sequencing), was developed.
• PRIME constructs a cell correspondence network and adjusts gene expression estimates using local subnetworks of similar cell types.
• The method was validated on synthetic and eight real-world scRNA-seq datasets.

Main Results:

• PRIME effectively reduces dropout effects in scRNA-seq data.
• The method demonstrably improves the quality of data visualization.
• PRIME enhances the accuracy of clustering analysis and aids in discovering noise-hidden gene expression patterns.

Conclusions:

• PRIME offers a robust solution for addressing dropout effects in scRNA-seq data.
• The imputation method leads to more reliable transcriptomic profiles.
• PRIME facilitates deeper biological insights from scRNA-seq experiments by uncovering subtle gene expression patterns.