scMoC

scMoC integrates scRNA-seq and scATAC-seq data to improve clustering of single cells by using RNA-guided imputation to reduce sparsity in scATAC-seq profiles.

Key Features:

• Imputation Strategy: Uses RNA-guided imputation leveraging less-sparse scRNA-seq measurements from the same cell to mitigate high sparsity in scATAC-seq data.
• Cross-Modal Clustering: Performs independent clustering in scRNA-seq and scATAC-seq domains and integrates those clustering results to combine information from gene expression and chromatin accessibility.
• Biologically Meaningful Clusters: Produces clusters supported by both RNA and scATAC-seq signals to reflect biological cell populations.

Scientific Applications:

• Cellular heterogeneity analysis: Enables multi-omic characterization of cellular heterogeneity by jointly leveraging gene expression and chromatin accessibility.
• Complex tissues and developmental processes: Applicable to studies of complex tissues or developmental systems where scRNA-seq and scATAC-seq provide complementary views of cell states.

Methodology:

Integrates scRNA-seq and scATAC-seq data from the same cell; applies RNA-guided imputation to enhance sparsity-limited scATAC-seq profiles; derives individual clusterings from each modality and merges them into integrated clusters.