Authors
Rahul Jamdade, Mariam Alsallani, Khawla Al Shaer, Sanjay Gairola, Tamer Mahmoud, Hatem A Shabana
Published in
Journal, genetic engineering & biotechnology. Volume 24. Issue 3. Pages 100716. Epub Jun 30, 2026.
Abstract
Halophytes represent a unique group of salt-tolerant plants that are critical to the ecology of arid and semi-arid environments, yet taxonomic resolution of some taxa remains challenging due to morphological plasticity and limited molecular data. This study employed a multilocus DNA barcoding approach using two chloroplast markers, rbcL and matK, and the nuclear ribosomal marker ITS2 to assess species-level resolution among halophytic plants from the United Arab Emirates (UAE). A total of 137 field-collected samples representing 38 species were processed, generating 373 high-quality barcode sequences across the three loci. Because PCR amplification and sequencing success varied among markers, not all samples yielded usable sequences for every locus. Consequently, the final marker-specific datasets comprised 132 sequences for rbcL, 125 for matK, and 116 for ITS2. To improve representation of species sampled by one or two individuals, 36 additional sequences were retrieved from GenBank, resulting in 409 sequences analyzed in total. The rbcL showed the highest amplification and sequencing success but the lowest species-level discrimination. In contrast, matK and ITS2 exhibited higher sequence variability and interspecific divergence, with ITS2 providing the strongest resolution for closely related taxa. Barcode gap and OTU-based ASAP analyses revealed limitations in resolving species within taxonomically complex genera, particularly Cyperus, Zygophyllum, and Tamarix. Supervised machine learning classifiers, especially Support Vector Machines, improved taxonomic assignment compared with traditional distance-based methods and achieved up to 97.7% classification accuracy. However, unresolved cases remained in closely related genera, reflecting biological constraints such as low interspecific divergence, recent divergence, hybridization, polyploidy, and limited barcode gaps. The present study demonstrates that multilocus barcoding combined with machine learning can improve species identification of UAE halophytes, while also highlighting the limitations of standard barcode loci in complex plant groups. The generated reference dataset contributes to the regional DNA barcode library and provides a molecular framework for biodiversity monitoring, conservation, and the sustainable utilization of salt-tolerant plant diversity in arid ecosystems.
PMID:
42749397
Bibliographic data and abstract were imported from PubMed on 17 Sep 2026.
Read full publication at:
Please sign in
to see all details.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 12
- Comments 0