scCCTR: An iterative selection-based semi-supervised clustering model for single-cell RNA-seq data.

Area of Science:

• Genomics and Bioinformatics
• Computational Biology
• Molecular Biology

Background:

• Single-cell RNA sequencing (scRNA-seq) is crucial for analyzing cellular heterogeneity.
• Existing cell clustering algorithms struggle with the high dimensionality and sparsity of scRNA-seq data.
• Improved computational methods are needed for accurate cell population identification.

Purpose of the Study:

• To develop a novel semi-supervised deep learning algorithm, scCCTR, for enhanced cell clustering in scRNA-seq data.
• To improve the accuracy and effectiveness of cell clustering and visualization.
• To address the limitations of existing clustering methods in handling complex scRNA-seq datasets.

Main Methods:

• Developed scCCTR, a novel semi-supervised classification algorithm for scRNA-seq data.
• Implemented an iterative selection module to identify high-confidence cells and optimize feature representations.
• Utilized a semi-supervised classification module with a Transformer neural network and multi-head attention for precise clustering.

Main Results:

• scCCTR demonstrated superior performance compared to established cell clustering methods on real datasets.
• The algorithm achieved higher accuracy and effectiveness in both cell clustering and visualization.
• Iterative selection and Transformer network integration led to improved clustering precision.

Conclusions:

• scCCTR offers a significant advancement in scRNA-seq data analysis for cell clustering.
• The novel approach effectively handles data challenges, leading to more reliable identification of cell populations.
• scCCTR provides a powerful tool for researchers studying cellular heterogeneity and diversity.