Nanopore signals of the DNAformer study

This repository contains the complete dataset of raw Nanopore signals supporting the publication: Bar-Lev, D., Orr, I., Sabary, O., Etzion T., & Yakkobi, E. Scalable and robust DNA-based storage via coding theory and deep learning. Nat Mach Intell, 7, 639–649 (2025). https://www.nature.com/articles/s42256-025-01003-z. Dataset description The repository includes the raw Nanopore signals in two formats: binned and original (unordered). Binned format: Each text file corresponds to a specific cluster and is named according to the cluster’s 12-base index (see publication for details). The file contains: All basecalled reads assigned to that cluster Their associated read IDs (unique identifiers of the signals) The corresponding FAST5 file in which each signal can be found Original (unordered) format: The raw signals from the pilot experiments are provided directly in this repository.The raw Nanopore signals of the test datasets are available separately in the following Zenodo repository:https://zenodo.org/records/13896773.