scHiCTools

scHiCTools provides computational analysis of single-cell Hi-C (scHi-C) sequencing data to process sparse contact maps and characterize three-dimensional chromatin organization at single-cell resolution.

Key Features:

• Data screening: Two methods to screen and validate single-cell Hi-C datasets.
• Smoothing techniques: Linear convolution, random walk, and network enhancement to refine contact maps and mitigate sparsity.
• Pairwise similarity calculation: Inner Product, HiCRep, and Selfish measures to compute pairwise similarities between cells.
• Dimensionality reduction: Multidimensional Scaling (MDS), t-Distributed Stochastic Neighbor Embedding (t-SNE), and Potential of Heat-diffusion for Affinity-based Trajectory Estimation (PHATE) for embedding scHi-C data into lower-dimensional Euclidean spaces.
• Clustering algorithms: scHiCluster, k-means, and spectral clustering to group cells by chromatin interaction profiles.
• Visualization: Functions to visualize cell embeddings in two-dimensional and three-dimensional scatter plots.

Scientific Applications:

• Cell-cycle dynamics: Analysis of cell-cycle progression and ordering using scHi-C embeddings and similarity measures.
• Cellular differentiation: Investigation of chromatin architecture changes across differentiation trajectories.
• Tissue heterogeneity: Identification and characterization of distinct cellular subpopulations or states within heterogeneous tissues.

Methodology:

Implemented in Python3.