Related Experiment Video
Updated: May 26, 2026
06:24
Multiplexed Analysis of Retinal Gene Expression and Chromatin Accessibility Using scRNA-Seq and scATAC-Seq
Published on: March 12, 2021
Quantifying Cross-Modal Association Confidence for Single-Cell RNA-ATAC Integration.
Tomoya Furutani1,2, Hongkai Ji1
1Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, 21212, USA.
Biorxiv : the Preprint Server for Biology
|May 25, 2026
Summary
We introduce the Cross-modality Link Confidence (CLIC) score to quantify gene expression and chromatin accessibility concordance. This score improves the integration of single-cell RNA-seq and ATAC-seq data, enhancing biological insights.
Area of Science:
- Genomics
- Computational Biology
- Bioinformatics
Background:
- Single-cell and spatial omics technologies generate vast datasets, but most measure only one molecular layer.
- Integrating separately profiled single-cell RNA-seq (scRNA-seq) and ATAC-seq (scATAC-seq) data is challenging due to variable correlations between gene expression and chromatin accessibility.
- Low-confidence gene-peak associations can reduce the accuracy of integrative computational methods.
Purpose of the Study:
- To develop a quantitative measure for the concordance between gene expression and chromatin accessibility in single-cell data.
- To introduce a novel feature selection strategy that leverages this concordance measure to improve cross-modal data integration.
- To enhance the robustness and biological interpretability of integrated single-cell omics datasets.
Main Methods:
- Developed the Cross-modality Link Confidence (CLIC) score using diverse single-cell multiome datasets from the ENCODE project.
- Quantified the empirical concordance between gene expression and nearby chromatin accessibility.
- Implemented a hybrid feature selection strategy combining highly variable genes with high-CLIC genes.
Main Results:
- CLIC scores provide reliable prior confidence estimates for gene-peak associations across modalities.
- The hybrid feature selection strategy significantly improves the integration of gene expression and chromatin accessibility data.
- The proposed approach demonstrated consistent performance across various public single-cell and spatial datasets and integration frameworks.
Conclusions:
- The CLIC score offers a robust method for assessing gene-regulatory element relationships in single-cell genomics.
- Integrating gene expression and chromatin accessibility data using CLIC-informed feature selection enhances analytical power and biological discovery.
- This work advances the computational toolkit for multimodal single-cell data analysis.
