مؤسسه انتشارات دانشگاه تهران علوم گیاهان زراعی ایران 2008-4811 57 1 2026 03 21 Genetic Diversity in Backcross Inbred Rice Lines Derived from Hashemi تنوع ژنتیکی در جمعیت لاین‌های اینبرد برنج حاصل از تلاقی برگشتی با رقم هاشمی 67 84 105737 10.22059/ijfcs.2025.402487.655160 FA سیده سهیلا زربافی سازمان تحقیقات، آموزش و ترویج کشاورزی، موسسه تحقیقات برنج کشور، رشت، ایران مریم حسینی چالشتری سازمان تحقیقات، آموزش و ترویج کشاورزی، موسسه تحقیقات برنج کشور، رشت، ایران Journal Article 2025 09 24 Introduction. Rice (Oryza sativa L.) is one of the world’s most important cereal crops, serving as a staple food for billions of people and providing a primary source of income for many agricultural communities. It is cultivated in over 100 countries, with Asia accounting for the majority of production and consumption. Rice contributes substantially to daily caloric and protein intake, supplying approximately 60% of daily caloric intake and nearly 50% of dietary protein in many regions (FAO, 2018). With the increasing global population, rising food demand, and climate variability, developing high-yielding, high-quality, and environmentally adaptable rice varieties has become essential for global food security and sustainable agriculture. Grain yield is a complex polygenic trait influenced by multiple morphological, physiological, and quality characteristics, which complicates direct selection in breeding programs (Debsharma et al., 2020; Nath, 2015). Understanding the relationships among yield and related traits is therefore critical for effective breeding. The exploration of genetic and phenotypic diversity within rice germplasm provides opportunities to identify superior genotypes and develop improved varieties, providing valuable resources for breeding programs (Thomson et al., 2007). Evaluating this variability using biometric parameters such as genetic and phenotypic variation, heritability, and expected genetic advance allows breeders to select traits with the greatest potential impact on yield and performance (Begna & Teressa, 2024). Multivariate statistical analyses, including principal component and cluster analyses, are effective tools for quantifying genetic diversity, revealing hidden patterns among traits, and guiding the selection of superior genotypes (Sharifi, 2018). These methods reduce data dimensionality, clarify complex trait interactions, and help prioritize traits for breeding programs, enabling more efficient improvement of rice varieties. In this context, the present study aimed to evaluate a population of rice genotypes in terms of agronomic and yield-related traits using multivariate statistical methods, with the goal of identifying genotypes possessing optimal combinations of traits for use in future breeding programs. Materials and Methods. A total of 144 backcross inbred lines (BILs, BC2F4) derived from a cross between Hashemi (recurrent parent, with superior cooking quality) and IR67418-110-32222 (donor parent, contributing desirable agronomic traits) were evaluated. The study aimed to combine the high-quality traits of Hashemi with improved morphological and yield-related characteristics from the donor parent. The experiment was conducted under field conditions in a randomized complete block design with three replications. Seedlings were transplanted at a spacing of 20 × 20 cm, and standard agronomic practices, including fertilization, irrigation, and pest and weed management, were applied. Ten quantitative traits were measured following the Standard Evaluation System (SES) for rice (IRRI, 2002): Grain yield, panicles per plant, spikelets per panicle, filled grains per panicle, 1,000-grain weight, plant height, panicle length, flag leaf area, days to 50% flowering, and days to maturity. Data were analyzed using SAS 9 (SAS Institute, 2002) for analysis of variance and mean comparisons, SPSS 24 (IBM SPSS Statistics, 2016) for correlation, factor, and cluster analyses, and Python 3.13 (Python Software Foundation, 2025) for heatmap visualization of trait associations. Superior and diverse genotypes were identified for potential use in future breeding programs. Results and Discussion. Descriptive statistics revealed wide variation among the BIL genotypes for all evaluated traits, including plant height, panicle length, panicles per plant, spikelets per panicle, flag leaf area, 1,000-grain weight, filled grains per panicle, and grain yield. The broad ranges observed for flowering and maturity times indicated the presence of early-, medium-, and late-duration genotypes within the population. Frequency distributions for most agronomic and yield-related traits were continuous and near-normal, suggesting polygenic inheritance and sufficient quantitative variation for effective selection. Analysis of variance showed significant differences among genotypes for all traits. Combined with the high heritability estimates, this indicated that much of the observed variation was genetically controlled. Correlation analysis revealed significant positive associations between grain yield and traits such as panicles per plant, flag leaf area, spikelets per panicle, filled grains per panicle, and panicle length, highlighting their value as indirect selection criteria. In contrast, the correlations between flowering traits and grain yield were small and non-significant, likely due to environmental conditions during the late growth stages and the stronger contribution of yield-component traits to final productivity. Factor analysis extracted four major components explaining 76.97% of the total variation, corresponding to growth and phenology, reproductive capacity, yield-defining traits, and grain physical characteristics. Cluster analysis grouped the genotypes into two groups: Group I contained high-yielding lines with superior values for major yield components, whereas Group II, which included the recurrent parent Hashemi, exhibited lower performance for yield-related traits. The cophenetic correlation coefficient of 0.88 confirmed the robustness of the clustering pattern. Overall, the results indicated that grain yield in the BIL population is governed by multiple interacting traits, and traits such as panicle length, filled grains per panicle, and flag leaf area are key indicators for identifying superior lines for breeding programs. Conclusion. The findings of this study demonstrated that the backcross-derived inbred population developed from Hashemi and IR67418-110-32222 provides a valuable genetic resource for improving both grain yield and quality in rice. Multivariate analyses confirmed that grain yield resulted from the combined influence of multiple morphological and reproductive traits rather than any single attribute, underscoring the need for multi-trait selection strategies. The clear genetic variation revealed through variance analysis, factor structure, and clustering further indicated the presence of distinct groups of superior genotypes that can serve as promising parents in future breeding programs. Overall, the study highlighted the importance of integrating morphological traits, yield components, and grain characteristics when selecting elite genotypes. These results provided practical guidance for breeders aiming to develop high-performing and high-quality rice cultivars adapted to diverse environments and support the continued use of multivariate approaches to accelerate breeding efficiency. برنج یکی از مهم‌ترین غلات دنیا و غذای اصلی جمعیت وسیعی از مردم است؛ از این رو شناسایی و بهره‌گیری از تنوع ژنتیکی ژنوتیپ‌ها در بهبود عملکرد و کیفیت محصول اهمیت زیادی دارد. مطالعه حاضر به ارزیابی تنوع ژنتیکی ۱۴۴ لاین اینبرد حاصل از تلاقی برگشتی رقم هاشمی و لاین IR67418-110-32222 پرداخت و ژنوتیپ‌ها از نظر صفات مورفولوژیکی و عملکردی مرتبط با هدف اصلاحی مورد بررسی قرار گرفتند. آزمایش در قالب طرح بلوک‌های کامل تصادفی با سه تکرار و اندازه‌گیری ده صفت مورفولوژیکی و عملکردی انجام شد. نتایج نشان داد که جمعیت مورد مطالعه از تنوع ژنتیکی قابلتوجهی برخوردار است و تمامی صفات مورد بررسی تفاوت‌های معنی‌دار ژنتیکی داشتند (p<0.01). عملکرد دانه بین 43/1 تا 42/6 تن در هکتار با میانگین 45/4 تن در هکتار متغیر بود. چهار عامل اصلی شناسایی‌شده، 97/76 درصد از واریانس کل صفات را تبیین کردند و تجزیه خوشه‌ای، ژنوتیپ‌ها را به دو گروه متمایز تقسیم کرد. ژنوتیپ‌های دارای طول خوشه بلندتر، تعداد دانه پر بیشتر و مساحت برگ پرچم وسیع‌تر، عملکرد بالاتری داشتند. بهطور کلی، این مطالعه نشان داد که استفاده از روش‌های چندمتغیره ابزاری مؤثر برای شناسایی ژنوتیپ‌ها با ترکیب مطلوب صفات هدف اصلاحی و انتخاب والدین مناسب در برنامه‌های اصلاحی است. جمعیت مورد بررسی می‌تواند به عنوان منبع ارزشمند ژنتیکی برای توسعه ارقام برنج با عملکرد بالا و کیفیت مطلوب مورد استفاده قرار گیرد.

https://ijfcs.ut.ac.ir/article_105737_1235027dd9a986d62af4e9b3ddca5a19.pdf