ScLinear predicts protein abundance at single-cell resolution.

Area of Science:

• Biomedical research
• Computational biology
• Machine learning in genomics

Background:

• Single-cell multi-omics technologies provide high-resolution biological insights.
• Predicting protein abundance from RNA expression is crucial for understanding cellular function.
• Current methods for this task can be computationally intensive.

Purpose of the Study:

• To introduce scLinear, a novel linear regression-based method.
• To predict single-cell protein abundance using RNA expression data.
• To offer a computationally efficient and accurate alternative to existing approaches.

Main Methods:

• Developed scLinear, a linear regression model.
• Trained and validated the model on single-cell RNA sequencing and protein data.
• Evaluated scLinear's performance against state-of-the-art methodologies.

Main Results:

• scLinear demonstrates high accuracy in predicting protein abundance.
• The method is significantly more efficient than current state-of-the-art approaches.
• scLinear generalizes well to unseen single-cell and spatial transcriptomics data.

Conclusions:

• scLinear provides an interpretable, efficient, and accurate solution for predicting protein levels from RNA data.
• The study advocates for considering simpler, faster methods in computational biology.
• scLinear advances the application of machine learning in single-cell multi-omics research.