NGS-Integrator

NGS-Integrator integrates multiple genome-wide next-generation sequencing (NGS) datasets by converting input tracks into complements of the minimum Bayes' factor and computing joint probabilities per genomic position to produce aggregated tracks for identifying regulatory elements and functional genomic domains.

Key Features:

• Minimum Bayes' Factor Integration: Computes the minimum Bayes' factor from the density of sequence reads mapped across the genome and expresses values as complements (cMBF) ranging between 0 and 1.
• Data Transformation: Transforms and standardizes heterogeneous input tracks onto the common cMBF scale to enable comparability across datasets.
• Joint Probability Calculation: Calculates the joint probability at each genomic position across all integrated cMBF-transformed tracks to generate an aggregated data track.
• Efficiency and Scalability: Uses time- and memory-efficient computation suitable for handling large-scale genome-wide NGS datasets.

Scientific Applications:

• Aggregate Data Tracks from Replicates: Produces genome-wide aggregated data tracks from multiple experimental replicates to enhance signal robustness.
• Integration of Complementary Experimental Data: Combines tracks from different experimental types, such as assays identifying enhancers or other regulatory elements, to refine feature identification.
• Downstream Regulatory Domain Analysis: Supports downstream analyses to identify functional regulatory domains and has been applied to real NGS datasets including examples from the mouse ENCODE database and gene transcription regulation studies.

Methodology:

Input tracks are transformed to complements of the minimum Bayes' factor (cMBF); joint probabilities are computed for each genomic position across transformed tracks; the final aggregated integrated track is output for downstream analysis.