Phenoflow-ML: A next-generation phenotyping framework to capture ML-based phenotypes

This manuscript presents Phenoflow-ML, a next-generation phenotyping framework that offers novel and powerful capabilities for ML-based phenotyping. This infrastructure is open-access and extensible, and its development was motivated by an important limitation of the state-of-the-art phenotyping too...

Full description

Saved in:
Bibliographic Details
Main Author: Antonio Lopez-Martinez-Carrasco
Format: Article
Language:English
Published: Elsevier 2025-09-01
Series:SoftwareX
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2352711025002833
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This manuscript presents Phenoflow-ML, a next-generation phenotyping framework that offers novel and powerful capabilities for ML-based phenotyping. This infrastructure is open-access and extensible, and its development was motivated by an important limitation of the state-of-the-art phenotyping tools: the lack of support for phenotypes based on machine learning. Some advantages of Phenoflow-ML are (a) its modular design, which improves code reuse and implementing new functionality, (b) the use of CWL, permitting the generation of portable and standard ML-based phenotype definitions for easy sharing between practitioners and institutions and enhancing research reproducibility, (c) the extensive documentation provided, and (d) the solid base of users and continuous growth since the initial versions. Furthermore, Phenoflow-ML has a significant impact in both research and practical areas, as it offers a powerful technical solution to define, represent, and share ML-based phenotype definitions in a clear, structured, and standard way.
ISSN:2352-7110