Just Add Data: automated predictive modeling for knowledge discovery and feature selection.

Autor: Tsamardinos I; JADBio Gnosis DA S.A., Science and Technology Park of Crete, GR-70013, Heraklion, Greece. tsamard.it@gmail.com.; Department of Computer Science, University of Crete, Heraklion, Greece. tsamard.it@gmail.com.; Institute of Applied and Computational Mathematics, Foundation for Research and Technology, Hellas, N. Plastira 100, Vassilika Vouton, Heraklion, GR-70013, Greece. tsamard.it@gmail.com., Charonyktakis P; JADBio Gnosis DA S.A., Science and Technology Park of Crete, GR-70013, Heraklion, Greece., Papoutsoglou G; JADBio Gnosis DA S.A., Science and Technology Park of Crete, GR-70013, Heraklion, Greece.; Department of Computer Science, University of Crete, Heraklion, Greece., Borboudakis G; JADBio Gnosis DA S.A., Science and Technology Park of Crete, GR-70013, Heraklion, Greece., Lakiotaki K; Department of Computer Science, University of Crete, Heraklion, Greece., Zenklusen JC; National Cancer Institute, National Institutes of Health, Bethesda, MD, USA., Juhl H; Chief Executive Officer, Indivumed Group, Hamburg, Germany., Chatzaki E; Laboratory of Pharmacology, Medical School, Democritus University of Thrace, Alexandroupolis, Greece.; Institute of Agri-food and Life Sciences, Hellenic Mediterranean University Research Centre, Crete, Greece., Lagani V; JADBio Gnosis DA S.A., Science and Technology Park of Crete, GR-70013, Heraklion, Greece.; Institute of Chemical Biology, Ilia State University, Tbilisi, Georgia.
Jazyk: angličtina
Zdroj: NPJ precision oncology [NPJ Precis Oncol] 2022 Jun 16; Vol. 6 (1), pp. 38. Date of Electronic Publication: 2022 Jun 16.
DOI: 10.1038/s41698-022-00274-8
Abstrakt: Fully automated machine learning (AutoML) for predictive modeling is becoming a reality, giving rise to a whole new field. We present the basic ideas and principles of Just Add Data Bio (JADBio), an AutoML platform applicable to the low-sample, high-dimensional omics data that arise in translational medicine and bioinformatics applications. In addition to predictive and diagnostic models ready for clinical use, JADBio focuses on knowledge discovery by performing feature selection and identifying the corresponding biosignatures, i.e., minimal-size subsets of biomarkers that are jointly predictive of the outcome or phenotype of interest. It also returns a palette of useful information for interpretation, clinical use of the models, and decision making. JADBio is qualitatively and quantitatively compared against Hyper-Parameter Optimization Machine Learning libraries. Results show that in typical omics dataset analysis, JADBio manages to identify signatures comprising of just a handful of features while maintaining competitive predictive performance and accurate out-of-sample performance estimation.
(© 2022. The Author(s).)
Databáze: MEDLINE
Externí odkaz: