pyBedGraph

pyBedGraph provides high-performance computation of summary statistics from genomic signal tracks in bedGraph and bigWig formats to quantify coverage and signal within genomic intervals for analyses such as ChIP-seq, ChIA-PET, DHS, ATAC-seq, and RNA-seq.

Key Features:

• File format support: Reads bedGraph text and bigWig binary coverage files for genomic signal analysis.
• Summary statistics: Computes mean, approximate mean, maximum, minimum, coverage, and standard deviation of signal within specified intervals.
• Approximate computation: Provides an approximate-mean option for faster, lower-cost estimation of interval means.
• Performance benchmarking: Demonstrated an average 260-fold speed increase over pyBigWig across 12 datasets (ChIP-seq, ATAC-seq, RNA-seq, ChIA-PET) and can compute exact means for one million regions in ~0.26 seconds and approximate means in <0.12 seconds on a standard laptop.
• Optimized implementation: Partly implemented in Cython to improve execution speed for large-scale analyses.

Scientific Applications:

• ChIP-seq and ChIA-PET signal quantification: Quantifies transcription factor binding intensity and protein-DNA interaction signals within genomic regions.
• ATAC-seq and DHS chromatin accessibility profiling: Summarizes chromatin accessibility signals across defined intervals.
• RNA-seq coverage analysis: Computes coverage and signal summaries for transcriptomic regions.
• High-throughput region-level summaries: Enables rapid computation of summary statistics for large sets of genomic intervals in high-throughput studies.

Methodology:

Reads bedGraph and bigWig files and performs fast statistical computations (mean, approximate mean, maximum, minimum, coverage, standard deviation) on specified genomic intervals; implementation is partly in Cython.