Z-GENIE: a user-friendly R/Shiny resource for predicting Z-DNA forming regions in DNA

Z-DNA is a left-handed DNA conformation with a zigzag backbone whose formation depends on base composition, modifications, and environmental factors. Although energetically unfavorable, Z-DNA has been implicated in both normal physiology and disease. The Z-Hunt algorithm predicts Z-DNA potential from thermodynamic principles, but its command-line interface and plain-text outputs limit adoption by users without coding expertise.

We introduce Z-GENIE, an R/Shiny GUI that automates Z-Hunt execution, parses its output, and presents interactive visualizations. Z-GENIE accepts FASTA files, NCBI accession IDs, or manual sequences and produces CSV and BED summaries compatible with genomic browsers. In benchmarks on small to medium genomes (< 20 Mb), Z-Hunt completes in minutes and the full Z-GENIE pipeline (data retrieval, parsing, visualization) finishes in under five minutes. For large genomes (> 50 Mb), Z-Hunt may require up to two hours, whereas Z-GENIE's parsing and BED-file export take < 2 min. In a human ADAM12 case study, Z-GENIE reproduced a published Z-score (3.0 × 10^7) and uncovered orientation-dependent Z-DNA clusters. Another case study compared predictions for Z-DNA in the rice genome (Oryza sativa) with experimental ZIP-Seq and CUT&Tag data; this study highlights the complementarity between in silico and in vivo approaches.

By encapsulating Z-Hunt within an intuitive GUI and offering flexible inputs and downstream-ready outputs, Z-GENIE democratizes genome-wide Z-DNA analysis. Its rapid performance and advanced visualization features should broaden exploration of Z-DNA's roles in health and disease.