A mini-review of single-cell Hi-C embedding methods

Area of Science:

• Genomics and Bioinformatics
• Computational Biology
• Molecular Biology

Background:

• Single-cell Hi-C (scHi-C) provides insights into 3D genome organization within individual nuclei.
• Analyzing scHi-C data requires advanced computational methods, particularly for handling sparse contact maps.
• Embedding techniques are crucial for dimensionality reduction and pattern extraction in scHi-C data.

Purpose of the Study:

• To systematically review existing computational methods for scHi-C data embedding.
• To evaluate the capabilities of these methods in normalization and imputation of sparse scHi-C data.
• To provide a benchmarking analysis comparing embedding techniques and their impact on cell clustering.

Main Methods:

• Systematic examination of scHi-C embedding methodologies.
• Assessment of normalization and imputation strategies within embedding techniques.
• Comprehensive benchmarking of embedding methods and their clustering performance.

Main Results:

• Identified and categorized various scHi-C embedding methods.
• Evaluated the effectiveness of normalization and imputation in improving scHi-C data quality.
• Benchmarked clustering performance across different embedding techniques.

Conclusions:

• ScHi-C embedding methods are essential for extracting meaningful biological patterns from 3D genome data.
• Normalization and imputation are critical steps for addressing data sparsity and enhancing interpretability.
• This review offers a practical guide for selecting appropriate scHi-C embedding tools for 3D genome organization studies.