Z-Calling: a tool for A/Z (2,6-diaminopurine) base calling and dZ-DNA detection using PacBio HiFi reads.

Area of Science:

• Genomics
• Biotechnology
• Bioinformatics

Background:

• The natural occurrence of 2,6-diaminopurine (Z) as an adenine (A) substitute in bacteriophage genomes presents evolutionary insights and biotechnological opportunities.
• Current limitations in detecting Z-DNA and distinguishing A/Z bases hinder research progress.
• Accurate sequencing of Z-DNA is crucial for understanding its biological roles and applications.

Purpose of the Study:

• To develop a reliable computational tool for identifying Z-DNA and differentiating A/Z bases from sequencing data.
• To enable single-nucleotide resolution of Z/A discrimination without additional sample processing.
• To advance the accessibility and accuracy of Z-DNA sequencing.

Main Methods:

• Development of Z-Calling, a machine learning tool utilizing PacBio Circular Consensus Sequencing (CCS) kinetic signal changes.
• Analysis of sequence context-dependent kinetic signals associated with Z/A substitution.
• Implementation of an A/Z base-calling module within the Z-Calling framework.

Main Results:

• Z-Calling reliably detects Z-DNA in mixed samples, even at ~1% Z-DNA content.
• The A/Z base-calling module achieves high performance with AUC scores of 0.942-0.952.
• The tool directly analyzes PacBio CCS reads, eliminating the need for further sample preparation.

Conclusions:

• Z-Calling significantly improves the ability to sequence and analyze Z-DNA.
• This advancement facilitates broader applications of Z-DNA in biotechnology and evolutionary studies.
• The freely available tool enhances accessibility for researchers in the field.