GEEES

GEEES infers cell-specific gene-enhancer interactions from multi-modal single-cell transcriptome and chromatin accessibility data and evaluates performance against benchmark datasets from bulk genomic experiments.

Key Features:

• Single-cell gene-enhancer inference: GEEES infers gene-enhancer associations at the individual cell level rather than at the level of grouped cells, preserving regulatory heterogeneity among cells.
• Multi-modal data integration: GEEES processes simultaneous single-cell transcriptome and chromatin accessibility data.
• Benchmark evaluation: GEEES was evaluated alongside multivariate regression-based alternatives and state-of-the-art methods using a large number of benchmark datasets derived from bulk genomic experiments.
• Distance-based improvement: Incorporating gene-enhancer distance into the analysis markedly improves performance.
• Cell representation learning: GEEES provides enhanced cell representation learning for downstream analysis.

Scientific Applications:

• Transcriptional regulation research: GEEES enables study of gene-enhancer interactions that are central to transcriptional regulation in a cell type-specific manner.
• Regulatory heterogeneity analysis: GEEES allows investigation of regulatory differences among individual cells rather than averaged cell populations.
• Method benchmarking: GEEES provides a framework for comprehensive assessment of current approaches for inferring gene-enhancer interactions from multi-modal single-cell data.

Methodology:

GEEES uses multi-modal single-cell transcriptome and chromatin accessibility data with incorporated gene-enhancer distance information; specific algorithmic details are not explicitly stated in the sources.