chroGPS

chroGPS represents similarities between epigenetic factors and genetic elements by projecting their epigenetic states into interpretable 2D and 3D reference maps for joint visualization of genomic and epigenomic data.

Key Features:

• Multidimensional scaling: Uses multidimensional scaling to project similarity relationships into 2D and 3D reference maps.
• chroGPS(factors): Creates maps that illustrate functional similarities between epigenetic factors based on their epigenetic states and genomic distributions.
• chroGPS(genes): Produces maps depicting the epigenetic states of genes and integrates gene expression and other functional information.
• Distance metrics and adjustment procedures: Applies distance metrics and adjustment procedures to mitigate systematic biases during data integration.
• Integration of heterogeneous datasets: Integrates genomic and epigenomic datasets obtained from different technologies or genetic backgrounds.
• Statistical robustness: Emphasizes statistical robustness to capture functional properties relevant for genomic interpretation.

Scientific Applications:

• Visualization of epigenetic relationships: Visualizes functional similarities and relationships among epigenetic factors using chroGPS(factors).
• Mapping gene epigenetic states: Maps epigenetic states of genes and incorporates gene expression and other functional data with chroGPS(genes).
• Cross-technology and cross-background integration: Enables integration and comparison of heterogeneous datasets from different sequencing technologies or genetic backgrounds while mitigating biases.
• Model organism demonstration: Applied to modENCODE genomic distribution data for multiple epigenetic factors in Drosophila melanogaster.
• Hypothesis generation: Captures functional properties of genomic elements to aid formulation of testable hypotheses.

Methodology:

Uses multidimensional scaling to generate 2D/3D reference maps from similarity or distance measures and applies distance metrics and adjustment procedures to mitigate systematic biases during data integration.