MOGEN

MOGEN converts chromosomal contact data into three-dimensional (3D) genome models to characterize the spatial organization of genomes.

Key Features:

• 3D reconstruction: Converts chromosomal contact data, particularly Hi-C, into three-dimensional genome models.
• Intra- and inter-chromosomal contacts: Generalizes a 3D chromosome reconstruction method to accommodate both intra- and inter-chromosomal Hi-C contacts.
• Robustness to noisy data: Handles noisy and inconsistent inter-chromosomal contact data.
• Validation with synthetic and experimental data: Validates models using synthetic datasets derived from polymer worm-like chain models and yeast genomes.
• Ensemble modeling: Generates ensemble 3D structural models from Hi-C data, including models of the human B-cell genome.
• Detection of spatial features: Recapitulates structural features such as chromosome compartmentalization, chromosome territories, co-localization of small chromosomes, interactions involving elongated or telomere regions, and dynamic orientations among chromosomes.

Scientific Applications:

• Genome function studies: Studying how 3D genome conformation relates to gene regulation, DNA replication, and genome methylation.
• Whole-genome 3D reconstruction: Reconstructing whole-genome 3D structures from Hi-C data including inter-chromosomal contacts.
• Structural organization analysis: Analyzing chromosome compartmentalization and chromosome territories in human and yeast genomes.
• Telomere and co-localization studies: Investigating co-localization of small chromosomes and interactions among telomere or elongated chromosome regions.
• Ensemble-based pattern assessment: Producing ensemble 3D structural models to examine structural patterns across models.

Methodology:

Converts chromosomal contact data from Hi-C into 3D models by generalizing a 3D chromosome reconstruction method to accommodate both intra- and inter-chromosomal Hi-C contacts; validation used synthetic datasets derived from polymer worm-like chain models and yeast genomes and ensemble models were generated for the human B-cell genome.