Dr. Hussein Abdullah Ahmed, a lecturer at the College of Agriculture, University of Kirkuk, has published a joint scientific research paper with researchers from the Department of Field Crops, Faculty of Agriculture, Ankara University / Türkiye; the Central Research Institute for Field Crops, Ministry of Agriculture / Türkiye; the Department of Mathematical and Statistical Methods, Poznań University of Life Sciences / Poland; the Department of Field Crops, Faculty of Agriculture, Necmettin Erbakan University / Türkiye; the Department of Agricultural Biotechnology, Faculty of Agriculture, Iğdır University / Türkiye; and the Department of Genetics, Plant Breeding and Seed Production, Wrocław University of Environmental and Life Sciences / Poland.
The paper was published in the international journal Euphytica, indexed in both Clarivate and Scopus databases.
The lecturer explained that the study highlights the limited genetic diversity in modern barley, which poses a major constraint to improving productivity. Therefore, 445 six-row barley genotypes from the Osman Tosun Gene Bank in Türkiye were analyzed by evaluating 22 agro-morphological traits (11 qualitative and 11 quantitative).
He added that principal component analysis (PCA) revealed that four components explained 72.86% of the total variation, with the first two accounting for 52.45%. Based on similarities, the genotypes were clustered into seven groups, with groups 5, 6, and 7 characterized by high-yield traits such as early maturity, number of grains per spike, and thousand-grain weight.
The study also compared three machine learning models — XGBoost, MARS, and Gaussian Processes — to predict the harvest index, where XGBoost outperformed the others, explaining 99.8% of the variation.
The findings further emphasized the importance of integrating machine learning techniques into barley breeding programs to enhance yield through precise prediction of desirable traits.
