ClementPla/DNAi: DNAi - Release

Read full text Summarise with VIRAT
Authors: ClementPla
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

DNAi v1.0.0 — Initial public release First stable release of DNAi, an open-source tool for automated segmentation, enumeration, and measurement of DNA replication fibers in fluorescence microscopy images. This release accompanies the manuscript accepted at Nucleic Acids Research. Features Deep learning segmentation of first and second analog labels (IdU/CldU), with ensemble inference and test-time augmentation (TTA) Automatic fiber classification by replication structure (ongoing forks, bidirectional origins, terminations) Error detection classifier to filter false positives Junction disentanglement for crossing and branching fibers Batch processing of whole-slide images with automated measurements and charts Cross-platform support for Zeiss (.czi), DeltaVision (.dv), TIFF, PNG, and JPEG formats Multiple interfaces: Streamlit GUI, Python API, Docker, and Google Colab Requirements Python 3.10+ CUDA-capable GPU recommended Links Documentation & usage: see README on github https://github.com/ClementPla/DNAi Docker image: clementpla/dnafiber on Docker Hub https://hub.docker.com/r/clementpla/dnafiber Google Colab notebook included: https://colab.research.google.com/github/ClementPla/DNAi/blob/main/DNAi.ipynb

Fields of Study & Subjects

Cell Image Analysis Techniques Medical Image Segmentation Techniques Image Processing Techniques and Applications

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
26 March 2026
Language
EN