Machine Learning Approach to Identify and Evaluate the High protein-yielding Varieties of Glycine max
Jordan Journal of Biological Sciences · 01.06.2026 · 273
DOI: 10.54319/jjbs/190210 ↗Kurz gefasst
Fachartikel in der Zeitschrift Jordan Journal of Biological Sciences, online erschienen am 01.06.2026.
Abstract (Originalsprache)
Soybeans are valued worldwide as legumes with exceptional health benefits, high nutritional value, and protein content.Though well-studied in many regions, there is limited information on the nutritional status of indigenous soybeans cultivated in the Uttarakhand region of the western Himalayan range in India, despite their prominence in the traditional food system.The present study aimed to assess protein content in indigenous soybean germplasms and use a machine learning algorithm to classify the nutritionally superior soybean samples.50 samples were collected from different regions of the hilly areas of the state and evaluated for protein content using the Kjeldahl method, followed by an HPLC analysis to ascertain the amino acid content in germplasms with high protein content.Machine learning method-the Decision Tree algorithm (J48) in WEKA (Waikato Environment for Knowledge Analysis) software was used to analyse the attributes like location, seed traits, temperature, and nutrient content.The ten-fold cross-validation achieved 94 % accuracy in classifying germplasms according to the protein content, as shown in the confusion matrix.The study showed that maximum protein content in the analyzed samples ranges from 36.96% to 43.56%, mostly in samples collected at an elevation of 345 m -1021 m, and 1699 m to 2044 m, latitudinal range of 28° to 30.9°, and longitudes 77.5° to 79.52°.HPLC analysis confirmed histidine as the most prominent amino acid, with the highest amount being 0.688 g/100g.
Zitieren
APA
(2026). Machine Learning Approach to Identify and Evaluate the High protein-yielding Varieties of Glycine max. Jordan Journal of Biological Sciences, 273. https://doi.org/10.54319/jjbs/190210
Vancouver
Machine Learning Approach to Identify and Evaluate the High protein-yielding Varieties of Glycine max. Jordan Journal of Biological Sciences. 2026:273. doi:10.54319/jjbs/190210
BibTeX
@article{euvimed2026,
title = {Machine Learning Approach to Identify and Evaluate the High protein-yielding Varieties of Glycine max},
journal = {Jordan Journal of Biological Sciences},
year = {2026},
pages = {273},
doi = {10.54319/jjbs/190210}
}
EUVIMED