MAPS

MAPS identifies significant long-range chromatin interactions from PLAC-seq and HiChIP datasets by correcting systematic biases in contact frequencies to map protein-anchored contacts.

Key Features:

• Zero-truncated Poisson regression: Employs a zero-truncated Poisson regression framework to model contact counts and account for count distributions without zeros.
• Bias correction and normalization: Normalizes chromatin contact frequencies and removes systematic biases inherent in PLAC-seq and HiChIP data.
• Protein-anchored interaction detection: Identifies significant interactions anchored at genomic regions bound by specific proteins, including transcription factors and histone marks.
• Compatibility with PLAC-seq and HiChIP: Designed to analyze datasets that combine Hi-C genome-wide contact mapping with ChIP-enrichment signals (PLAC-seq and HiChIP).
• Performance: Demonstrates superior performance compared to existing software when analyzing multiple PLAC-seq and HiChIP datasets.

Scientific Applications:

• Chromatin architecture mapping: High-resolution mapping of long-range chromatin interactions to study genome organization and regulation.
• Protein-centric interaction analysis: Characterization of chromatin contacts anchored at transcription factors and histone modification sites.
• Comparative dataset analysis: Analysis and comparison of interaction profiles across multiple PLAC-seq or HiChIP experiments.

Methodology:

Applies a zero-truncated Poisson regression to model and normalize contact counts, remove systematic biases, and identify significant interactions anchored at protein-bound genomic regions.