RoboCOP

RoboCOP computes genome-wide probabilistic occupancy landscapes for nucleosomes and transcription factors by integrating nucleotide sequence with chromatin accessibility data such as MNase-seq and ATAC-seq.

Key Features:

• Multivariate state space model: Uses a multivariate state space model to infer chromatin occupancy dynamics.
• Data integration: Integrates nucleotide sequence information with chromatin accessibility data to inform occupancy inference.
• Supported assays: Applies to MNase-seq and ATAC-seq data and is described in the context of DNase-seq accessibility data.
• DBF scope: Models occupancy for nucleosomes and transcription factors, collectively referred to as DNA-binding factors (DBFs).
• Joint inference: Jointly computes probabilistic occupancy scores for hundreds of TFs and nucleosomes across the genome.
• Genome-wide output: Produces genome-wide probabilistic occupancy landscapes for nucleosomes and TFs.
• Yeast application: Has been applied to map nucleosomes and 150 transcription factors across the yeast genome.
• Perturbation analysis: Enables analysis of chromatin dynamics under conditions such as cadmium stress to reveal changes in transcriptional regulation.
• Comparative performance: Demonstrated superior predictive capability relative to existing methods in reported applications.

Scientific Applications:

• Chromatin occupancy mapping: Mapping the protein-binding landscape of nucleosomes and transcription factors genome-wide.
• Joint TF and nucleosome analysis: Simultaneous probabilistic scoring of hundreds of TFs alongside nucleosome occupancy to resolve factor-specific binding within accessibility data.
• Condition-specific chromatin dynamics: Investigating changes in chromatin occupancy and associated transcriptional regulation under conditions such as cadmium stress in yeast.
• Method comparison: Benchmarking occupancy inference performance against existing computational methods.

Methodology:

Integrates nucleotide sequence with chromatin accessibility data (MNase-seq, ATAC-seq, discussed relative to DNase-seq) using a multivariate state space model to compute genome-wide probabilistic occupancy landscapes for nucleosomes and transcription factors and to jointly score hundreds of DBFs.