Evaluating the Performance of Peak Calling Algorithms Available for Intracellular G-Quadruplex Sequencing

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• DNA G-quadruplexes (G4) are crucial non-canonical DNA structures involved in various biological processes.
• Antibody-dependent sequencing is vital for detecting intracellular G4s, requiring effective peak calling for data analysis.

Purpose of the Study:

• To systematically assess and compare existing peak calling algorithms for their suitability in analyzing intracellular DNA G4 sequencing data.
• To identify the most effective algorithms for accurate G4 detection in biological samples.

Main Methods:

• Seven representative peak calling algorithms were selected from 43 published methods.
• Performance was evaluated on six diverse intracellular G4 sequencing datasets using precision and recall metrics.
• Validation involved comparison with known G4 information and epigenetic signals.

Main Results:

• MACS2, PeakRanger, and GoPeaks were identified as the top-performing algorithms for intracellular G4 sequencing data analysis.
• Algorithm effectiveness was partly attributed to their read/fragment distribution models during hypothesis testing.
• The study provides a benchmark for selecting appropriate peak callers for G4 detection.

Conclusions:

• The study successfully identified optimal peak calling algorithms for intracellular DNA G4 analysis.
• Findings offer crucial guidance for researchers working with G4 sequencing data.
• This evaluation enhances the accuracy and reliability of G4 structure identification in biological studies.