Integrating morphophysiology, gene expression and machine learning to characterize salt and drought stress responses in dragon fruit.

Authors

Seher Toprak, Ömer Faruk Coşkun

Published in

Scientific reports. Jul 11, 2026. Epub Jul 11, 2026.

Abstract

Abiotic stresses are major environmental constraints that limit plant growth and productivity, particularly in arid and semi-arid regions where salinity and drought frequently occur simultaneously. In this study, the effects of salt stress (150 mM NaCl), PEG-induced drought stress (4% PEG-8000), and their combined application on the in vitro responses of dragon fruit (Hylocereus undatus) were evaluated at morphophysiological, molecular, and machine learning levels. The selected NaCl and PEG concentrations were used to impose moderate-to-severe salinity and osmotic stress conditions previously reported to affect pitaya growth and physiology. Cladode explants were exposed to stress treatments for 14 days, after which visual stress score, lateral shoot number, stem length, root length, stem diameter, fresh weight, dry weight, root number, and SPAD value were measured. In addition, the expression levels of five stress-responsive genes (HuERF1, HuTZF3, Cu/Zn-SOD, CAT, and NCED) were analyzed using RT-qPCR. Salt stress and, more prominently, the combined salt-drought treatment markedly impaired plant growth and physiological performance. Compared with the control, the combined treatment reduced root length by approximately 87%, fresh weight by 67%, root number by 54%, and SPAD value by 46%, indicating severe inhibition of root development, biomass accumulation, and chlorophyll-related performance. Gene expression analysis revealed strong induction of HuERF1 and Cu/Zn-SOD under salt-containing treatments, suggesting activation of stress-signaling and antioxidant defense responses, whereas NCED showed the highest upregulation under PEG-induced osmotic stress and combined stress, indicating the involvement of ABA-mediated osmotic adjustment. In contrast, CAT expression remained relatively stable across treatments. Morphological data were further analyzed using a Random Forest classifier, which distinguished stress treatments with an overall accuracy of approximately 73%. Feature importance analysis identified SPAD value, visual stress score, lateral shoot number, root number, and fresh weight as the most informative morphophysiological variables for treatment discrimination. Overall, these results demonstrate that salinity and drought, especially in combination, substantially affect growth performance and stress-responsive gene regulation in H. undatus, and highlight the potential of machine learning for identifying phenotypic signatures associated with abiotic stress responses.

PMID:
42436339
Bibliographic data and abstract were imported from PubMed on 12 Jul 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.