Next Article in Journal
A Non-Destructive System Using UVE Feature Selection and Lightweight Deep Learning to Assess Wheat Fusarium Head Blight Severity Levels
Previous Article in Journal
A New Method for Single-Plant Selection of Wheat Genotypes for Tolerance and Resistance to the Root-Lesion Nematode Pratylenchus thornei by Low-Density Sowing
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Agronomy
Volume 15
Issue 9
10.3390/agronomy15092050
agronomy-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Identification of Candidate Genes Related to SPAD Value Using Multi-Year Phenotypic Data in Rice Germplasms by Genome-Wide Association Study (GWAS)

by
Dong-Hyun Baek
1,†,
Tae-Heon Kim
1,2,†,
Chang-Ju Lee
1,
Jingli Gao
1,
Woo-Geun Park
1 and
Suk-Man Kim
1,*
1
Department of Ecological & Environmental System, Kyungpook National University, Sangju 37224, Republic of Korea
2
Institute of Agricultural Science and Technology, Kyungpook National University, Daegu 41566, Republic of Korea
*
Author to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Agronomy 2025, 15(9), 2050; https://doi.org/10.3390/agronomy15092050
Submission received: 22 July 2025 / Revised: 19 August 2025 / Accepted: 23 August 2025 / Published: 26 August 2025
(This article belongs to the Section Crop Breeding and Genetics)
Browse Figures
Versions Notes

Abstract

This study aimed to identify candidate genes associated with chlorophyll content in rice via genome-wide association studies (GWAS) and to develop molecular markers for the selection of genetic resources and breeding lines exhibiting high chlorophyll content. Measurement of the Soil and Plant Analysis Development (SPAD) values, indicative of chlorophyll content and photosynthetic potential, were measured in 198 rice genetic resources across three years under consistent nitrogen conditions. Nitrogen fertilizer (as urea) was applied at a rate of 90 kg N ha−1. After analyzing the multi-year SPAD data, genetic resources with the coefficient of variation (CV) value exceeding 20% were excluded, and the remaining 175 accessions were used for subsequent analyses. Population structure analysis using the principal component analysis (PCA) and phylogenetic methods confirmed clear genetic differentiation, supporting the reliability of the GWAS. A GWAS using 289,569 SNPs identified 17 significant loci, among which four quantitative trait loci (QTLs)—qSV3-1, qSV3-2, qSV6, and qSV7—explained over 20% of phenotypic variance. Analysis of their additive effects revealed distinct SPAD distributions among QTL combination groups, with accessions harboring all four QTLs exhibiting the highest values. Candidate gene analysis within ± 200 kb of lead SNPs identified Os03g079100 (OsUCL8), involved in photosynthesis, near qSV3-2. A derived cleaved amplified polymorphic sequence (dCAPS) marker was developed to differentiate alleles at this locus and validated via restriction digestion. These results provide key genetic insights into chlorophyll accumulation and offer molecular markers for breeding high-yielding rice cultivars with enhanced chlorophyll content. The results of this study are expected to contribute to the development of sustainable rice varieties by utilizing the developed markers and identified candidate genes to increase SPAD values, thereby enhancing nitrogen use efficiency, improving photosynthetic capacity, and ultimately increasing rice productivity.

1. Introduction

Rice (Oryza sativa L.) is one of the world’s three major staple crops and is cultivated in over 100 countries, serving as a primary food source for nearly half of the global population [1]. Notably, in Asia, rice contributes up to 50% of daily caloric intake [2] and ranks as the second most extensively grown crop worldwide [3]. During the 1970s, the Green Revolution dramatically transformed agricultural productivity through the adoption of high-yielding varieties, expansion of irrigation systems, and increased use of chemical fertilizers [4]. Notably, rice production heavily relies on nitrogen fertilization to sustain high yields. Nitrogen plays a central role as a key component of chlorophyll, influencing leaf color, plant growth, and crop productivity [5].
Leaf chlorophyll content is closely associated with photosynthetic capacity and serves as an important indicator of plant physiological status [6]. Chlorophyll content was found to be correlated with leaf nitrogen content [7,8], and notably, higher nitrogen content in rice plants suggested a potential association with nitrogen use efficiency [9,10]. Measurement of the Soil and Plant Analysis Development (SPAD) value is a simple, rapid, and non-destructive method for estimating relative chlorophyll content in leaves. SPAD values have shown strong correlations with actual chlorophyll concentrations in various crop species, including rice, maize, wheat, and soybean [11,12,13,14,15,16,17]. Thus, exploiting genes associated with the regulation of SPAD values offers a promising strategy to enhance photosynthetic performance and nitrogen use efficiency in rice, which in turn could support the breeding of cultivars suitable for low-nitrogen input systems and resilient under climate change.
However, SPAD values can be influenced by various factors, such as cultivar differences, growth stage, leaf thickness, measurement position, and leaf location [18,19,20,21,22,23].
In certain years or time points, SPAD values have exhibited strong correlations with nitrogen status; however, results can vary significantly across years or measurement periods, even at similar developmental stages. Thus, SPAD should be interpreted cautiously and primarily as an indicator of nitrogen deficiency [24]. Although extensive research has explored SPAD variation and leaf color responses to nitrogen fertilization, precise conclusions remain elusive due to genetic and environmental complexity. In natural conditions, plants are subject to diverse environmental stresses that interfere with photosynthesis and reduce yield potential. Since photosynthesis is highly sensitive to environmental factors [25], its measurement has become a vital method in plant stress research [26].
Additionally, studies on SPAD variation are limited by the number of available genetic resources and the difficulty in conducting year-to-year evaluations, highlighting the need for further investigation. In this study, we examined SPAD variation using diverse rice genetic resources. We employed a genome-wide association study (GWAS) to identify single nucleotide polymorphisms (SNPs) associated with SPAD values and their corresponding candidate genes. A total of 198 rice (Oryza sativa L.) genetic resources were used as experimental materials, and the SPAD value of the flag leaf for each material was evaluated using a SPAD-502 Plus chlorophyll meter (Konica Minolta Inc., Tokyo, Japan). The SNPs associated with SPAD values were found to increase the SPAD values of the genetic resources, and candidate genes related to photosynthetic efficiency were ultimately identified. Our findings may contribute to the development of improved rice cultivars with enhanced yield potential and nitrogen use efficiency, thereby supporting the stability and productivity of rice cultivation.

2. Materials and Methods

2.1. Plant Materials

A total of 198 rice (Oryza sativa L.) genetic resources representing diverse ecotypes were used as experimental materials in this study. The genetic resources included 114 japonica, 41 indica, 33 admixed, and 10 others (3 aromatic, 7 aus).

2.2. Field Experimental Conditions and Fertilization Practices

The rice genetic resources were cultivated from 2022 to 2024 at the Experimental Farm of the College of Agricultural and Life Sciences, Kyungpook National University, located in Hyoryeong-myeon, Gunwi-gun, Daegu, Korea. Planting density was uniformly maintained throughout the three-year period. Each treatment was laid out in a randomized complete block design (RCBD) with three replications per plot. Each plot for a genetic resource covered an area of 1.125 m2, consisting of a single line with 25 hills per line, at a spacing of 30 × 15 cm, with one plant established per hill. The genotype × replicate combination was treated as the experimental unit.
Nitrogen (N), phosphorus (P), and potassium (K) were applied individually as single fertilizers. The fertilizer application rates were 90 kg N ha−1, 45 kg P ha−1, and 75 kg K ha−1 for nitrogen, phosphorus, and potassium, respectively. Urea was applied as the source of nitrogen fertilizer. Nitrogen was split into basal, tillering, and panicle fertilization, accounting for 50%, 20%, and 30% of the total N rate, respectively. Potassium was divided at a 70:30 ratio between basal and panicle fertilization, while phosphorus was entirely applied as a basal dose. Basal fertilization was conducted three days prior to transplanting, tillering fertilization 14 days after transplanting, and panicle fertilization 30 days before heading. In 2022, the soil physicochemical properties of the experimental field were as follows: pH, 6.6; organic matter (OM), 29.9 g/kg; NH4+, 15.2 mg/kg; and NO3−, 16.0 mg/kg. In 2023, the values were pH, 5.5; OM, 38.9 g/kg; NH4+, 9.8 mg/kg; and NO3−, 3.0 mg/kg (data not shown). Irrigation of the experimental field was applied as needed. Other agronomic practices followed the standard cultivation methods of rice [27].

2.3. Phenotypic Evaluation

Leaf color of the rice genetic resources was measured using a SPAD-502 Plus chlorophyll meter, and the relative chlorophyll content was expressed as SPAD values [28]. Measurements were conducted on the middle portion of the flag leaf at the heading stage, with the leaf blade precisely positioned in the center of the device’s sample slot between 10:00 am and 3:00 pm on sunny days. A single reading point was taken for each leaf/plant. SPAD values were recorded across three replications, with five individual plants measured per replication annually. For each replicate, five SPAD readings per genotype were averaged after removing outliers based on the interquartile range (IQR).

2.4. High-Throughput SNP Genotyping

High-throughput SNP genotyping was performed using the KNU Axiom Oryza 580 K platform [29]. After hybridization and imaging with the GeneTitan MC instrument (Affymetrix, Santa Clara, CA, USA), SNPs were filtered based on quality criteria and selected for GWAS. In total, 289,569 SNPs meeting the thresholds of minor allele frequency (MAF) > 0.05 and missing genotype rate < 0.2 were included in the analysis.

2.5. Population Structure and Phylogenetic Analysis

Population structure analysis of the 175 rice genetic resources was performed using principal component analysis (PCA) in R. PCA plots were generated through the GAPIT (Genome Association and Prediction Integrated Tool, version 3) package to visualize genetic variation among rice genetic resources. To determine the optimal number of genetic clusters, an elbow plot analysis was conducted using the entire SNP dataset (289,569 markers), which revealed a distinct inflection point at K = 2 [30]. Phylogenetic relationships were inferred based on the selected SNPs using TASSEL version 5. Additionally, a circular phylogenetic tree was constructed in R to complement the analysis and illustrate genetic clustering across rice genetic resources.

2.6. GWAS Analysis and SNP Annotation

Genome-wide association analysis was conducted using version 3 of the GAPIT package in R. Five statistical models—BLINK (Bayesian-information and Linkage Disequilibrium Iteratively Nested Keyway), CMLM (Compressed Mixed Linear Model), FarmCPU (Fixed and Random Model Circulating Probability Unification), MLM (Mixed Linear Model), and MLMM (Multi-Locus Mixed Linear Model)—were employed. The mean SPAD values collected over a three-year period were used as phenotypic input data. The principal components (PCs) were used to control for population structure in GWAS. We used the PCA.total = 2 parameters for running the GAPIT package.
Manhattan and Q–Q plots were generated using the GAPIT package and RStudio. The statistical significance threshold was calculated as –log10(1/number of effective independent SNPs) for the identification of SNPs significantly associated with SPAD values. Lead SNPs were annotated according to trait specificity. SNPs that explained over 20% of phenotypic variation were designated as qSV (SPAD Value).

2.7. Candidate Gene Identification

Phenotypic variation associated with SPAD-related lead SNPs identified from the GWAS analysis was evaluated across rice genetic resources. Allelic differences in phenotype were assessed using t-tests across the genetic resources. Genomic positions of significantly associated SNPs were compared with known rice gene annotations located within a ±200 kb window [31], using databases such as the Rice Annotation Project Database (https://rapdb.dna.affrc.go.jp/, accessed on 25 February 2025) and Gramene (https://www.gramene.org/, accessed on 25 February 2025). Gene function information was retrieved using R packages.

2.8. Marker Development

To select SNPs, GWAS–QTL (quantitative trait loci)-colocalized candidate genes on chromosome 3 were examined. Primer3PLUS v3.3.0 software and assay design service provided by SnapGene software v6.0.2 were used to design the SNP markers [32,33].

2.9. Statistical Analysis

Statistical analyses were performed using R software (version 4.5.1; The R Foundation for Statistical Computing); t-tests, correlation analyses, and regression analyses were used to compare SPAD value distributions among rice genetic resources. Various R packages were utilized to facilitate data processing and visualization. Broad-sense heritability (h2) was calculated as the ratio of the genotypic variance to the phenotypic variance using the formula [34]:
h 2 = σ 2 g / σ 2 p h × 100
where h2 = broad sense heritability (%), σ2g = genotypic variance, and σ2ph = phenotypic variance.

3. Results

3.1. Annual Variation in SPAD Values

SPAD values of 198 rice genetic resources were measured over three consecutive years. In 2022, SPAD values ranged from 30.5 to 55.6, with a mean of 40.0 ± 5.3. In 2023, the range was 34.4 to 58.5, with a mean of 42.8 ± 4.6, while in 2024, values ranged from 28.0 to 54.0, with a mean of 40.0 ± 4.0 (Figure S1 and Table 1). Annual variation in SPAD values has been reported to reflect differences in crop growth stages and environmental conditions [35].
Among the 198 rice genetic resources, average SPAD values ranged from 32.2 to 55.3, with an overall mean of 40.9 ± 4.4. The CV ranged from 0.2% to 12.8% (Table 1 and Figure S2). Based on CV analysis, 23 genetic resources with inter-annual variability exceeding 8.3% were excluded. Consequently, the final dataset included 175 genetic resources, for which the three-year average SPAD values were used in subsequent analyses (Figure 1). For these 175 rice genetic resources, SPAD values under nitrogen treatment ranged from 32.4 to 55.3, with a mean of 41.1 ± 4.3.

3.2. Genetic Structure Revealed by Phylogenetic and PCA Analyses

The 175 rice genetic resources were classified into five genetic groups—japonica, indica, admixed, aus, and aromatic—with reference to Guo et al. [36]. The group composition included 104 japonica, 34 indica, 30 admixed, 5 aus, and 2 aromatic resources. Using the selected 289,569 SNPs, a phylogenetic tree was constructed to visualize genetic relationships among the resources (Figure 2).
PCA was performed to assess genetic diversity. The PCA results, generated using the GAPIT package, revealed that PC1 and PC2 accounted for 59.49% and 4.6% of the total genetic variation, respectively. PCA clustering was based on the five groups, allowing for evaluation of genetic distribution patterns across ecospecies.
To determine the optimal number of clusters, an elbow plot analysis was performed using the selected 289,569 SNPs. The within-cluster sum of squares (WSS) was visualized across different values of K, representing the number of clusters. A distinct inflection point was observed at K = 2, where the WSS value sharply declined [37]. This suggests that two clusters most effectively capture the overall genetic variation among the rice genetic resources.

3.3. SPAD Variation by Ecotype and Genotypic Grouping

SPAD values of 175 genetic resources were analyzed according to ecospecies classification. Genetic resources belonging to the indica group exhibited significantly lower SPAD values compared to other ecospecies (Figure 3). The genotypic classification of 175 genetic resources revealed two distinct clusters. A comparison of SPAD values between these groups was conducted using a t-test, which indicated a statistically significant difference at the 95% confidence level (p < 0.05).

3.4. SNP Dataset Utilized for GWAS Analysis

A total of 289,569 SNPs were identified from the 175 rice genetic resources through SNP-chip and next-generation sequencing (NGS) analysis. The number of SNPs per chromosome ranged from 16,551 to 37,852, with chromosome 12 containing the fewest SNPs and chromosome 1 the most. The average number of markers per chromosome was 24,131.
Chromosome sizes varied from 23.0 Mb (chromosome 9, the smallest) to 43.2 Mb (chromosome 1, the largest). SNP density per Mb also differed across chromosomes, with chromosome 12 showing the lowest density (591.11 SNPs/Mb) and chromosome 2 the highest (919.72 SNPs/Mb) (Figure S3). The average SNP density across all 12 chromosomes was 757.35 SNPs/Mb (Table S1).

3.5. Identification of SPAD-Associated SNPs Through GWAS

GWAS analysis was conducted using five statistical models—BLINK, CMLM, FarmCPU, MLM, and MLMM—implemented in the GAPIT package to identify SNPs associated with SPAD values across 175 rice genetic resources (Figure 4). The analysis identified 17 SNPs significantly associated with SPAD values. These SNPs are considered candidate quantitative trait nucleotides (QTNs) and may also represent QTLs that influence leaf chlorophyll content (Figure S4). From the Q–Q plots, GWAS analyses of the mean SPAD values across years using the BLINK, CMLM, FarmCPU, MLM, and MLMM models showed either minimal or no inflation (Figure 4).

3.6. SNPs Related to SPAD Value by GWAS Analysis

A total of 17 SNPs were found to be significantly associated with SPAD values. The number of SNPs identified in each model was as follows: BLINK (4), CMLM (2), FarmCPU (4), MLM (2), and MLMM (5). These SNPs were distributed across chromosomes 2, 3, 6, 7, 8, and 10. The –log10 (p) values ranged from 5.48 to 18.19. The phenotypic variation explained (PVE) by individual SNPs ranged from 0% to 51.24% (Table 2).

3.7. Selection of QTLs Based on PVE

To evaluate the phenotypic effects of lead SNPs identified through GWAS on SPAD values, SNPs with a PVE greater than 20% were selected. Variation in SPAD values was evaluated based on the presence or absence of lead SNPs. For each lead SNP, its presence was defined by the possession of the allele corresponding to a higher average SPAD value. Based on these results, SNPs suitable for candidate gene identification were prioritized. SNPs meeting these conditions were designated as =QTLs. The identified QTLs were located on chromosome 3 (qSV3-1, qSV3-2), chromosome 6 (qSV6), and chromosome 7 (qSV7) (Table 3).

3.8. Analysis of QTL Additive Effects

Four QTLs with PVE values greater than 20% showed additive effects on the SPAD values. These QTLs—qSV3-1, qSV3-2, qSV6, and qSV7—were located on chromosomes 3, 6, and 7 (Figure S5), and their respective alleles are summarized in Table 3. To assess the additive effects of these QTLs, rice genetic resources were grouped based on QTL combinations, resulting in six distinct groups: A (qSV3-1 + qSV3-2 + qSV6 + qSV7), B (qSV3-1 + qSV6 + qSV7), C (qSV3-1 + qSV7), D (qSV6 + qSV7), E (qSV6), and F (None) (Figure 5). The SPAD values ranged from 32.4 to 55.4, with a mean of 41.1 ± 4.3 across the population. Group A showed the highest SPAD values, ranging from 47.6 to 55.3 with a mean of 50.7 ± 2.7, followed by Group B (40.2–53.8, mean 45.6 ± 2.9), Group C (34.1–44.3, mean 38.7 ± 2.1), Group D (38.0–45.0, mean 40.8 ± 2.9), Group E (32.4–42.3, mean 37.9 ± 3.0), and Group F (33.7–44.5, mean 39.9 ± 2.8) (Table S2). These results suggest that the presence and combination of SPAD-associated QTLs contribute to distinct phenotypic patterns, with Group A exhibiting the strongest cumulative genetic effect on SPAD values.

3.9. Candidate Gene Identification Associated with the SPAD Value

To identify candidate genes, the genomic regions spanning ± 200 kb around each significant QTL (SNP) were examined using the Rice Annotation Project Database (https://rapdb.dna.affrc.go.jp/, accessed on 25 February 2025). Notably, a candidate gene related to photosynthesis, Os03g079100, was identified near the SPAD-associated SNP qSV3-2 on chromosome 3, within the 28.5–28.6 Mb region (Figure 6 and Table S3).

3.10. Determination of Allele Type of qSV3-2

To validate the distribution of the qSV3-2 allele within the 175 rice genetic resources, a derived cleaved amplified polymorphic sequence (dCAPS) marker targeting a specific SNP site was developed (Table S4). This marker was designed to differentiate accessions harboring the A allele (n = 9) from those carrying the G allele (n = 165), using the restriction enzyme Dde I for digestion. The remaining genetic resource was not included due to missing genotyping data. To confirm the marker’s discriminatory capacity, a PCR test was performed on eight representative resources—four possessing the A allele (Jizi1581, Vialone Nano, Binhae Col.1, and Jonong) and four with the G allele (Nipponbare, 93-11, Basmati 389, and Dular). All eight resources produced a uniform amplicon of 212 bp. Following Dde I enzyme digestion, only the resources containing the G allele exhibited fragment cleavage, whereas A-allele accessions remained intact (Figure 7). Genetic resources carrying the A allele exhibited a mean SPAD value of 50.7, whereas those carrying the G allele showed a mean value of 40.6 (Figure 8). These results demonstrate the marker’s efficacy in reliably distinguishing SNP alleles among diverse rice genotypes.

4. Discussion

4.1. Estimation of SPAD Value Variation

SPAD values are closely related to leaf chlorophyll content, which varies depending on nitrogen fertilization levels and is influenced by both genetic and environmental factors [38]. Consequently, genetic resources that show high year-to-year variation are likely to reflect not only environmental influence but also underlying genetic differences. To minimize the impact of fluctuating environmental conditions across years and better capture consistent genetic tendencies, mean SPAD values across multiple growing seasons were used. This approach provides a more reliable estimation of genetic patterns compared with single-year data. The CV serves as a relative measure of data variability and may fluctuate depending on dataset size and contextual information from prior research [39]. CV offers insight into the precision of experimental data, with a lower CV indicating higher precision and accuracy. Based on general standards for field experiments, a CV below 10% is considered low, 10–20% moderate, 20–30% high, and above 30% very high [40]. Greater precision leads to smaller differences between mean values. The interpretation of CV varies depending on the type of experiment, crop species, and traits measured—for example, varietal trials for rice yield typically show CVs of 6–8%, fertilizer experiments 10–12%, and pesticide or herbicide trials 13–15% [41].
Although SPAD measurements provide a rapid and non-destructive proxy for chlorophyll content, their use as an absolute estimator is subject to several limitations. Variations in leaf morphological traits, such as thickness, surface texture, and internal anatomy, can alter light transmittance and absorption, leading to discrepancies between SPAD values and actual chlorophyll concentration. Furthermore, measurement position along the leaf blade and the phenological stage at the time of sampling may significantly influence the readings. These factors underscore the importance of standardized measurement protocols and, when possible, complementary biochemical assays to validate SPAD-based estimations.

4.2. Insights into Population Structure

The confirmation of the population structure of rice genetic resources through PCA and phylogenetic tree analysis, as well as the distribution of groups according to ecotypes, provides information on the genetic background and structure for the identification of SNPs associated with SPAD values. This helps to reduce the risk of biased results arising from these factors [42,43,44]. The spatial distribution of groups within the PCA plot provided insights into the genetic diversity and population structure of rice genetic resources. It is noteworthy that while PCA clusters are formed based on structural patterns of genetic variation, phylogenetic tree construction relies on genetic distances and evolutionary relationships among SNPs. As a result, some discrepancies in group placement were observed between the two analyses.

4.3. Allelic Influence of Lead SNPs on SPAD Value

The SPAD value is a method for estimating the relative chlorophyll content in leaves. Chlorophyll content is closely related to photosynthetic capacity and is known to serve as an indicator for evaluating the physiological status of crops as well as enhancing photosynthetic efficiency. To identify candidate genes that determine or influence SPAD levels, four lead SNPs were selected.
Several previous studies have identified genetic loci associated with SPAD values and chlorophyll-related traits in rice. Takai et al. [45] identified a QTL for SPAD value on chromosome 4 through substitution mapping using a set of chromosome segment substitution lines (CSSLs). Wang et al. [46] detected 17 SNPs associated with chlorophyll content-related traits across chromosomes 1, 2, 3, 4, 5, 6, 7, and 12. Matsubara et al. [47] identified six QTLs related to SPAD values on chromosomes 1, 2, 3, and 4. Liu et al. [48] reported that the candidate gene Os03g058300 was significantly associated with SPAD value based on a GWAS. Honda et al. [49] identified two QTLs related to chlorophyll content and leaf photosynthesis on chromosomes 4 and 8 using GWAS. Meng et al. [50] detected two SNPs associated with SPAD values on chromosomes 7 and 11 using GWAS. In the present study, we identified four SNPs significantly associated with SPAD values through GWAS. Notably, qSV3-2 was located on chromosome 3, which overlaps with a previously reported QTL (28,527,366–30,114,777 bp) by Ishimaru et al. [51] for chlorophyll content (Figure S5). This finding reinforces the relevance of this genomic region in regulating chlorophyll-related traits in rice.
Phenotypic variation in SPAD values was compared between allele groups for each SNP across the rice genetic resources. When distinct phenotypic separation was observed between alleles, the corresponding SNPs were considered potentially useful for the development of DNA markers for marker-assisted selection (MAS) and further functional validation.

4.4. Implications of Additive QTL Effects on SPAD Variation and Marker Development

The average SPAD values of Groups C, D, E, and F were lower than the overall mean SPAD value of the 175 rice genetic resources, indicating that these QTL combinations did not exhibit a cumulative effect. In contrast, Groups A and B showed higher SPAD values than the population mean, with Group A—containing all four QTLs—displaying the highest SPAD values. Group B, which lacked only qSV3-2 compared with Group A, also showed elevated SPAD levels; however, due to the absence of a genotype carrying qSV3-2 alone, the individual effect of qSV3-2 could not be determined.
Using the lead SNP from qSV3-2, we developed a dCAPS marker that effectively distinguished the SNP alleles. The SPAD values associated with each allele were 50.7 for the A allele and 40.6 for the G allele, indicating that the A allele functions as a positive allele contributing to higher SPAD values. The dCAPS marker has potential utility in marker-assisted selection for rice breeding programs aimed at developing cultivars with higher SPAD values. The SNP markers identified in this study were derived from a rice core set representing specific genetic backgrounds. While these markers are promising candidates for improving chlorophyll content and potentially enhancing nitrogen use efficiency, their effectiveness in breeding populations with distinct genetic backgrounds may vary. Therefore, validation studies in diverse germplasm and multi-environment trials are necessary before large-scale deployment in marker-assisted selection programs. Such efforts would help ensure the robustness and broad applicability of the identified markers across different breeding programs.

4.5. Identification of Candidate Gene

A candidate gene, Os03g079100, was identified within the ±200 kb genomic region around qSV3-2 on chromosome 3. Os03g079100, OsUCL8, encodes a member of the phytocyanin protein family. Phytocyanins are copper-containing electron transport proteins found in chloroplasts and are closely involved in regulating photosynthesis, stress responses, and plant growth and development [52,53]. In plants, copper plays a critical role in photosynthesis, respiration, electron transport, oxidative stress management, and cell wall metabolism [54].
Our findings provide valuable insights for breeding rice cultivars with improved nitrogen use efficiency through the selection of alleles associated with enhanced chlorophyll content. The identified markers could be incorporated into marker-assisted selection programs to facilitate the development of cultivars capable of maintaining photosynthetic capacity, thereby contributing to sustainable rice production and environmental conservation.

5. Conclusions

The objective of this study was to identify candidate genes associated with chlorophyll content in rice via GWAS and to develop molecular markers to facilitate the selection of high-chlorophyll genetic resources and breeding lines. This study revealed substantial variation in SPAD values among 175 rice genetic resources under consistent nitrogen fertilization across three growing seasons. By integrating multi-year phenotypic data with high-resolution genomic analysis, we successfully identified four QTLs—qSV3-1, qSV3-2, qSV6, and qSV7—significantly associated with the SPAD value of the flag leaf at heading. Notably, the cumulative effect of these loci demonstrated a clear genetic contribution to elevated SPAD values, with Group A (carrying all four QTLs) exhibiting the highest photosynthetic potential.
Population structure analyses using PCA and phylogenetic methods further affirmed the genetic diversity present within the experimental panel, reinforcing the reliability of GWAS outcomes. The identification of OsUCL8, a putative photosynthesis-related gene located near qSV3-2, underscores the biological relevance of candidate loci and offers a promising target for future functional validation.
Collectively, our findings contribute to a more refined understanding of the genetic architecture underlying leaf color traits in rice and provide a molecular basis for chlorophyll biosynthesis and its association with rice productivity. This work lays the groundwork for developing high-yielding cultivars with enhanced photosynthesis-related physiological traits, thus advancing sustainable rice production under variable environmental conditions.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/agronomy15092050/s1, Table S1. Distribution of SNPs across the 12 rice chromosomes used in GWAS analysis in this study. Table S2. Evaluation of SPAD values with respect to epistasis of QTLs. Table S3. Identification of candidate genes associated with SPAD values within a 200 kb flanking region of lead SNPs. Table S4. Information on the SNP marker (dCAPS) for selecting SPAD values. Figure S1. Distribution of SPAD values in 198 rice genetic resources across three years (2022–2024) and their mean values. Figure S2. Scatter plot of 198 genetic resources showing the relationship between coefficient of variation (CV) and standard deviation (SD). Figure S3. Distribution of SNP density on each chromosome. Figure S4 Distribution of lead SNPs associated with SPAD values using 5 models (BLINK, CMLM, FarmCPU, MLM, and MLMM) in GWAS. Figure S5. Phenogram plot showing significant QTLs (−log10(p) > 5.46) associated with SPAD values.

Author Contributions

Conceptualization, S.-M.K.; methodology, T.-H.K. and S.-M.K.; software, D.-H.B.; validation, D.-H.B. and T.-H.K.; formal analysis, D.-H.B.; investigation, C.-J.L., J.G. and W.-G.P.; data curation, J.G. and W.-G.P.; writing—original draft preparation, D.-H.B.; writing—review and editing, T.-H.K. and S.-M.K.; visualization, C.-J.L. and D.-H.B.; funding acquisition, S.-M.K.; D.-H.B. and T.-H.K. equally contributed to this paper and should be regarded as co-first authors. All authors have read and agreed to the published version of the manuscript.

Funding

The author(s) declare that financial support was received for the research, authorship, and/or publication of this article. This research was supported by the Rural Development Administration (RDA) of South Korea, grant number RS-2022-RD010269.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors on request.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Koumeleh, A.S.; Sharmila, P.; Upretyl, D.; Saradhi, P.P. Impact of elevated CO2 on nutrient uptake of rice cultivars (Oryza sativa L.). Indian J. Crop Sci. 2007, 2, 87–90. [Google Scholar]
  2. Li, Y.; Xiao, J.; Chen, L.; Huang, X.; Cheng, Z.; Han, B.; Zhang, Q.; Wu, C. Rice functional genomics research: Past decade and future. Mol. Plant 2018, 11, 359–380. [Google Scholar] [CrossRef]
  3. Giri, D.; Dhital, M.; Chaudhary, B.; Pandey, R.; Bastakoti, B.; Shrestha, S. Effect of different nitrogen levels on yield and yield attributes of different rice varieties in DDSR condition at Kanchanpur, Nepal. Arch. Agric. Environ. Sci. 2022, 7, 310–317. [Google Scholar] [CrossRef]
  4. Haque, M.A.; Haque, M.M. Growth, yield and nitrogen use efficiency of new rice variety under variable nitrogen rates. Am. J. Plant Sci. 2016, 7, 612–622. [Google Scholar] [CrossRef]
  5. Fageria, N.K.; Baligar, V.C.; Jones, C.A. Growth and Mineral Nutrition of Field Crops; CRC Press: Boca Raton, FL, USA, 2010. [Google Scholar]
  6. Peng, J.; Feng, Y.; Wang, X.; Li, J.; Xu, G.; Phonenasay, S.; Luo, Q.; Han, Z.; Lu, W. Effects of nitrogen application rate on the photosynthetic pigment, leaf fluorescence characteristics, and yield of indica hybrid rice and their interrelations. Sci. Rep. 2021, 11, 7485. [Google Scholar] [CrossRef]
  7. Patra, B.; Jena, S.; Satapathy, M.; Panda, R.; Behera, S.; Sahu, S. Chlorophyll content and spad value of rice (Oryza sativa L.) under different sowing windows and nutrient management practices. Ann. Plant Soil Res. 2021, 23, 324–328. [Google Scholar] [CrossRef]
  8. Gholizadeh, A.; Saberioon, M.; Borůvka, L.; Wayayok, A.; Soom, M.A.M. Leaf chlorophyll and nitrogen dynamics and their relationship to lowland rice yield for site-specific paddy management. Inf. Process. Agric. 2017, 4, 259–268. [Google Scholar] [CrossRef]
  9. Liang, T.; Duan, B.; Luo, X.; Ma, Y.; Yuan, Z.; Zhu, R.; Peng, Y.; Gong, Y.; Fang, S.; Wu, X. Identification of high nitrogen use efficiency phenotype in rice (Oryza sativa L.) through entire growth duration by unmanned aerial vehicle multispectral imagery. Front. Plant Sci. 2021, 12, 740414. [Google Scholar] [CrossRef]
  10. Srikanth, B.; Subrahmanyam, D.; Sanjeeva Rao, D.; Narender Reddy, S.; Supriya, K.; Raghuveer Rao, P.; Surekha, K.; Sundaram, R.M.; Neeraja, C.N. Promising physiological traits associated with nitrogen use efficiency in rice under reduced N application. Front. Plant Sci. 2023, 14, 1268739. [Google Scholar] [CrossRef]
  11. Monje, O.A.; Bugbee, B. Inherent limitations of nondestructive chlorophyll meters: A comparison of two types of meters. HortScience 1992, 27, 69–71. [Google Scholar] [CrossRef]
  12. Reeves, D.; Mask, P.; Wood, C.; Delaney, D. Determination of wheat nitrogen status with a hand-held chlorophyll meter: Influence of management practices. J. Plant Nutr. 1993, 16, 781–796. [Google Scholar] [CrossRef]
  13. Turner, F.T.; Jund, M.F. Chlorophyll meter to predict nitrogen topdress requirement for semidwarf rice. Agron. J. 1991, 83, 926–928. [Google Scholar] [CrossRef]
  14. Wakiyama, Y. The relationship between SPAD values and leaf blade chlorophyll content throughout the rice development cycle. Jpn. Agric. Res. Q. JARQ 2016, 50, 329–334. [Google Scholar] [CrossRef]
  15. Zotarelli, L.; Cardoso, E.G.; Piccinin, J.L.; Urquiaga, S.; Boddey, R.M.; Torres, E.; Alves, B.J.R. Calibration of a Minolta SPAD-502 chlorophyll meter for evaluation of the nitrogen nutrition of maize. Pesqui. Agropecuária Bras. 2003, 38, 1117–1122. [Google Scholar] [CrossRef]
  16. He, H.; Liu, C.; Wu, Z.; Chen, M.; Qu, K.; Zhao, J.; Wang, Y.; Hu, Z.; Li, Q. Responses of Rice Photosynthesis and Yield to Elevated CO2 Concentrations: A Quantitative Analysis via Chlorophyll Fluorescence Technology. J. Soil Sci. Plant Nutr. 2024, 24, 102. [Google Scholar] [CrossRef]
  17. Kim, T.H.; Kim, S.M. Effects of SPAD value variations according to nitrogen application levels on rice yield and its components. Front. Plant Sci. 2024, 15, 1437371. [Google Scholar] [CrossRef]
  18. Hoel, B.O.; Solhaug, K.A. Effect of irradiance on chlorophyll estimation with the Minolta SPAD-502 leaf chlorophyll meter. Ann. Bot. 1998, 82, 389–392. [Google Scholar] [CrossRef]
  19. Huang, J.; He, F.; Cui, K.; Buresh, R.J.; Xu, B.; Gong, W.; Peng, S. Determination of optimal nitrogen rate for rice varieties using a chlorophyll meter. Field Crops Res. 2008, 105, 70–80. [Google Scholar] [CrossRef]
  20. Li, J.W.; Yang, J.P.; Li, D.S.; Fei, P.P.; Guo, T.T.; Ge, C.S.; Chen, W.Y. Chlorophyll meter’s estimate of weight-based nitrogen concentration in rice leaf is influenced by leaf thickness. Plant Prod. Sci. 2011, 14, 177–183. [Google Scholar]
  21. Peng, S.; Buresh, R.J.; Huang, J.; Yang, J.; Zou, Y.; Zhong, X.; Wang, G.; Zhang, F. Strategies for overcoming low agronomic nitrogen use efficiency in irrigated rice systems in China. Field Crops Res. 2006, 96, 37–47. [Google Scholar] [CrossRef]
  22. Thiyagarajan, T.; Geetha, S.; Balasubramanian, V. Assessing genotypic variation in N requirements of rice with a chlorophyll meter. Int. Rice Res. Notes 2000, 25, 23–24. [Google Scholar]
  23. Yang, H.; Yang, J.; Lv, Y.; He, J. SPAD values and nitrogen nutrition index for the evaluation of rice nitrogen status. Plant Prod. Sci. 2014, 17, 81–92. [Google Scholar] [CrossRef]
  24. Kim, M.; Shin, J.; Lee, B. Applicability of vegetation index and SPAD reading to nondestructive diagnosis of rice growth and nitrogen nutrition status. Korean J. Crop Sci. 2005, 50, 369–377. [Google Scholar]
  25. Kalaji, H.M.; Carpentier, R.; Allakhverdiev, S.I.; Bosa, K. Fluorescence parameters as early indicators of light stress in barley. J. Photochem. Photobiol. B Biol. 2012, 112, 1–6. [Google Scholar] [CrossRef]
  26. Kalaji, H.M.; Jajoo, A.; Oukarroum, A.; Brestic, M.; Zivcak, M.; Samborska, I.A.; Cetner, M.D.; Łukasik, I.; Goltsev, V.; Ladle, R.J. Chlorophyll a fluorescence as a tool to monitor physiological status of plants under abiotic stress conditions. Acta Physiol. Plant. 2016, 38, 1–11. [Google Scholar] [CrossRef]
  27. Rural Development Administration. Manual for Standard Evaluation Method in Agricultural Experiment and Research; Rural Development Administration (RDA): Suwon, Republic of Korea, 2003. [Google Scholar]
  28. Kim, D.S.; Yoon, Y.H.; Shin, J.C.; Kim, J.K.; Kim, S.D. Varietal Difference in Relationship between SPAD Value and Chlorophyll and Nitrogen Concentration in Rice Leaf. Korean J. Crop Sci. 2002, 47, 263–267. [Google Scholar]
  29. Kim, K.-W.; Nawade, B.; Nam, J.; Chu, S.-H.; Ha, J.; Park, Y.-J. Development of an inclusive 580K SNP array and its application for genomic selection and genome-wide association studies in rice. Front. Plant Sci. 2022, 13, 1036177. [Google Scholar] [CrossRef]
  30. Saputra, D.M.; Saputra, D.; Oswari, L.D. Effect of distance metrics in determining k-value in k-means clustering using elbow and silhouette method. In Proceedings of the Sriwijaya International Conference on Information Technology and Its Applications (SICONIAN 2019), Palembang, Indonesia, 16 November 2019; pp. 341–346. [Google Scholar]
  31. Yu, J.; Zao, W.; He, Q.; Kim, T.-S.; Park, Y.-J. Genome-wide association study and gene set analysis for understanding candidate genes involved in salt tolerance at the rice seedling stage. Mol. Genet. Genom. 2017, 292, 1391–1403. [Google Scholar] [CrossRef]
  32. Koressaar, T.; Remm, M. Enhancements and modifications of primer design program Primer3. Bioinformatics 2007, 23, 1289–1291. [Google Scholar] [CrossRef]
  33. Untergasser, A.; Cutcutache, I.; Koressaar, T.; Ye, J.; Faircloth, B.C.; Remm, M.; Rozen, S.G. Primer3—New capabilities and interfaces. Nucleic Acids Res. 2012, 40, e115. [Google Scholar] [CrossRef]
  34. Akinwale, M.; Gregorio, G.; Nwilene, F.; Akinyele, B.; Ogunbayo, S.; Odiyi, A. Heritability and correlation coefficient analysis for yield and its components in rice (Oryza sativa L.). Afr. J. Plant Sci. 2011, 5, 207–212. [Google Scholar]
  35. Piekielek, W.P.; Fox, R.H.; Toth, J.D.; Macneal, K.E. Use of a chlorophyll meter at the early dent stage of corn to evaluate nitrogen sufficiency. Agron. J. 1995, 87, 403–408. [Google Scholar] [CrossRef]
  36. Guo, D.; Li, Y.; Lu, H.; Zhao, Y.; Kurata, N.; Wei, X.; Wang, A.; Wang, Y.; Zhan, Q.; Fan, D.; et al. A pangenome reference of wild and cultivated rice. Nature 2025, 642, 662–671. [Google Scholar] [CrossRef]
  37. Badapanda, K.; Mishra, D.P.; Salkuti, S.R. Agriculture data visualization and analysis using data mining techniques: Application of unsupervised machine learning. TELKOMNIKA (Telecommun. Comput. Electron. Control) 2022, 20, 98–108. [Google Scholar] [CrossRef]
  38. Yuan, Z.; Cao, Q.; Zhang, K.; Ata-Ul-Karim, S.T.; Tian, Y.; Zhu, Y.; Cao, W.; Liu, X. Optimal Leaf Positions for SPAD Meter Measurement in Rice. Front. Plant Sci. 2016, 7, 719. [Google Scholar] [CrossRef]
  39. Steel, R.G.D.; Torrie, J.H. Principles and Procedures of Statistics, 3rd ed.; McGraw-hill: New York, NY, USA, 1997. [Google Scholar]
  40. Pimentel-Gomes, F. Curso de Estatística Experimental; FEALQ: Piracicaba, Brazil, 2009; Volume 15, ISBN 9788571330559. [Google Scholar]
  41. Gomez, K.A.; Gomez, A.A. Statistical Procedures for Agricultural Research, 2nd ed.; John Wiley & Sons: New York, NY, USA, 1984. [Google Scholar]
  42. Maji, A.; Shaibu, A. Application of principal component analysis for rice germplasm characterization and evaluation. J. Plant Breed. Crop Sci. 2012, 4, 87–93. [Google Scholar] [CrossRef]
  43. Wang, Q.; Li, X.; Chen, H.; Wang, F.; Li, Z.; Zuo, J.; Fan, M.; Luo, B.; Feng, P.; Wang, J. Mapping combined with principal component analysis identifies excellent lines with increased rice quality. Sci. Rep. 2022, 12, 5969. [Google Scholar] [CrossRef]
  44. Nachimuthu, V.V.; Robin, S.; Sudhakar, D.; Raveendran, M.; Rajeswari, S.; Manonmani, S. Evaluation of rice genetic diversity and variability in a population panel by principal component analysis. Indian. J. Sci. Technol. 2014, 7, 1555–1562. [Google Scholar] [CrossRef]
  45. Takai, T.; Kondo, M.; Yano, M.; Yamamoto, T. A quantitative trait locus for chlorophyll content and its association with leaf photosynthesis in rice. Rice 2010, 3, 172–180. [Google Scholar] [CrossRef]
  46. Wang, Q.; Xie, W.; Xing, H.; Yan, J.; Meng, X.; Li, X.; Fu, X.; Xu, J.; Lian, X.; Yu, S. Genetic architecture of natural variation in rice chlorophyll content revealed by a genome-wide association study. Mol. Plant 2015, 8, 946–957. [Google Scholar] [CrossRef]
  47. Matsubara, K.; Yamamoto, E.; Kobayashi, N.; Ishii, T.; Tanaka, J.; Tsunematsu, H.; Yoshinaga, S.; Matsumura, O.; Yonemaru, J.-i.; Mizobuchi, R. Improvement of rice biomass yield through QTL-based selection. PLoS ONE 2016, 11, e0151830. [Google Scholar] [CrossRef]
  48. Liu, S.; Xiong, Z.; Zhang, Z.; Wei, Y.; Xiong, D.; Wang, F.; Huang, J. Exploration of chlorophyll fluorescence characteristics gene regulatory in rice (Oryza sativa L.): A genome-wide association study. Front. Plant Sci. 2023, 14, 1234866. [Google Scholar] [CrossRef]
  49. Honda, S.; Imamura, A.; Seki, Y.; Chigira, K.; Iwasa, M.; Hayami, K.; Nomura, T.; Ohkubo, S.; Ookawa, T.; Nagano, A.J. Genome-wide association study of leaf photosynthesis using a high-throughput gas exchange system in rice. Photosynth. Res. 2024, 159, 17–28. [Google Scholar] [CrossRef]
  50. Meng, B.; Wang, T.; Luo, Y.; Xu, D.; Li, L.; Diao, Y.; Gao, Z.; Hu, Z.; Zheng, X. Genome-wide association study identified novel candidate loci/genes affecting lodging resistance in rice. Genes 2021, 12, 718. [Google Scholar] [CrossRef]
  51. Ishimaru, K.; Yano, M.; Aoki, N.; Ono, K.; Hirose, T.; Lin, S.; Monna, L.; Sasaki, T.; Ohsugi, R. Toward the mapping of physiological and agronomic characters on a rice function map: QTL analysis and comparison between QTLs and expressed sequence tags. Theor. Appl. Genet. 2001, 102, 793–800. [Google Scholar] [CrossRef]
  52. Abdel-Ghany, S.E.; Pilon, M. MicroRNA-mediated systemic down-regulation of copper protein expression in response to low copper availability in Arabidopsis. J. Biol. Chem. 2008, 283, 15932–15945. [Google Scholar] [CrossRef]
  53. Shao, M.; Feng, Y.; Yang, S.; Feng, T.; Zeng, F.; Lu, S.; Ma, Z.; Chen, B.; Mao, J. Molecular evolution of Phytocyanin gene and analysis of expression at different coloring periods in apple (Malus domestica). BMC Plant Biol. 2024, 24, 374. [Google Scholar] [CrossRef]
  54. Marschner, H. Mineral Nutrition of Higher Plants, 2nd ed.; Academic Press: London, UK, 1995. [Google Scholar]
Agronomy 15 02050 g001
Figure 1. Distribution of mean SPAD values in 175 rice accessions evaluated over three years. (a) Histogram of mean SPAD values across three years. (b) Violin plots showing the distribution of mean SPAD values; the white dot represents the median, the vertical black bar represents the interquartile range, and thin black lines denote the lower and upper ranges. Arrows with smooth ends indicate mean ± SD. (c) Comparison of flag leaf coloration based on SPAD values.
Figure 1. Distribution of mean SPAD values in 175 rice accessions evaluated over three years. (a) Histogram of mean SPAD values across three years. (b) Violin plots showing the distribution of mean SPAD values; the white dot represents the median, the vertical black bar represents the interquartile range, and thin black lines denote the lower and upper ranges. Arrows with smooth ends indicate mean ± SD. (c) Comparison of flag leaf coloration based on SPAD values.
Agronomy 15 02050 g001
Agronomy 15 02050 g002
Figure 2. Phylogenetic tree and principal component analysis (PCA) of 175 rice genetic resources. (a) Phylogenetic tree constructed from 289,569 SNPs, classifying the accessions into five genetic groups: japonica (n = 104), indica (n = 34), admixed (n = 30), aus (n = 5), and aromatic (n = 2). (b) PCA plot of the 175 rice genetic resources, with PC1 and PC2 explaining 59.49% and 4.60% of the total genetic variation, respectively. (c) Elbow plot for determining the optimal number of clusters (K); the red dot indicates the elbow point.
Figure 2. Phylogenetic tree and principal component analysis (PCA) of 175 rice genetic resources. (a) Phylogenetic tree constructed from 289,569 SNPs, classifying the accessions into five genetic groups: japonica (n = 104), indica (n = 34), admixed (n = 30), aus (n = 5), and aromatic (n = 2). (b) PCA plot of the 175 rice genetic resources, with PC1 and PC2 explaining 59.49% and 4.60% of the total genetic variation, respectively. (c) Elbow plot for determining the optimal number of clusters (K); the red dot indicates the elbow point.
Agronomy 15 02050 g002
Agronomy 15 02050 g003
Figure 3. Comparison of SPAD values in 175 rice accessions. (a) Boxplot of SPAD values across subpopulations. (b) Boxplot of SPAD values grouped by PCA-based clusters. The black horizontal line indicates the median, boxes represent the interquartile range, and whiskers denote the upper and lower quartiles. Small circles indicate outliers. Lowercase letters indicate significant differences among groups based on one-way ANOVA (p < 0.05, Duncan’s test). Asterisks (*) denote significant differences at the 0.05 probability level according to the t-test.
Figure 3. Comparison of SPAD values in 175 rice accessions. (a) Boxplot of SPAD values across subpopulations. (b) Boxplot of SPAD values grouped by PCA-based clusters. The black horizontal line indicates the median, boxes represent the interquartile range, and whiskers denote the upper and lower quartiles. Small circles indicate outliers. Lowercase letters indicate significant differences among groups based on one-way ANOVA (p < 0.05, Duncan’s test). Asterisks (*) denote significant differences at the 0.05 probability level according to the t-test.
Agronomy 15 02050 g003
Agronomy 15 02050 g004
Figure 4. Manhattan and quantile–quantile (Q–Q) plots of genome-wide association studies (GWAS) for SPAD values. The upper horizontal solid line indicates the genome-wide significance threshold (−log10(p) = 5.46, p = 3.5 × 10−6), and the dashed line indicates the suggestive significance threshold. Dots above the threshold represent lead SNPs identified across rice chromosomes. The Q–Q plot shows the expected versus observed −log10(p) of each marker (blue dots). The red line is a benchmark for perfect fit to the expected −log10(p). The gray-shaded area represents the 95% confidence interval for the Q–Q plot under the null hypothesis of no association between the SNP and the trait.
Figure 4. Manhattan and quantile–quantile (Q–Q) plots of genome-wide association studies (GWAS) for SPAD values. The upper horizontal solid line indicates the genome-wide significance threshold (−log10(p) = 5.46, p = 3.5 × 10−6), and the dashed line indicates the suggestive significance threshold. Dots above the threshold represent lead SNPs identified across rice chromosomes. The Q–Q plot shows the expected versus observed −log10(p) of each marker (blue dots). The red line is a benchmark for perfect fit to the expected −log10(p). The gray-shaded area represents the 95% confidence interval for the Q–Q plot under the null hypothesis of no association between the SNP and the trait.
Agronomy 15 02050 g004
Agronomy 15 02050 g005
Figure 5. Comparison of SPAD values for detected QTL combinations. Capital letters (A–F) represent different QTL combinations: A, qSV3-1 + qSV3-2 + qSV6 + qSV7; B, qSV3-1 + qSV6 + qSV7; C, qSV3-1 + qSV7; D, qSV6 + qSV7; E, qSV6; F, none. Different lowercase letters indicate significant differences in SPAD values at p = 0.05 according to Duncan’s multiple range test. The red dashed line denotes the overall mean SPAD value (41.1 ± 4.3).
Figure 5. Comparison of SPAD values for detected QTL combinations. Capital letters (A–F) represent different QTL combinations: A, qSV3-1 + qSV3-2 + qSV6 + qSV7; B, qSV3-1 + qSV6 + qSV7; C, qSV3-1 + qSV7; D, qSV6 + qSV7; E, qSV6; F, none. Different lowercase letters indicate significant differences in SPAD values at p = 0.05 according to Duncan’s multiple range test. The red dashed line denotes the overall mean SPAD value (41.1 ± 4.3).
Agronomy 15 02050 g005
Agronomy 15 02050 g006
Figure 6. Exploration of putative candidate genes located within a ±200 kb region of qSV3-2. Red triangles indicate SNPs significantly associated with SPAD values as identified by GWAS. Surrounding gene models represent annotated genes within the ±200 kb flanking region. Among them, OsUCL8 (on chromosome 3), highlighted in red, was considered a potential candidate gene due to its putative functional relevance.
Figure 6. Exploration of putative candidate genes located within a ±200 kb region of qSV3-2. Red triangles indicate SNPs significantly associated with SPAD values as identified by GWAS. Surrounding gene models represent annotated genes within the ±200 kb flanking region. Among them, OsUCL8 (on chromosome 3), highlighted in red, was considered a potential candidate gene due to its putative functional relevance.
Agronomy 15 02050 g006
Agronomy 15 02050 g007
Figure 7. Genotyping of qSV3-2 (SNP) using a dCAPS marker. (a) Electrophoresis of PCR products from each rice genetic resource using the dCAPS marker. (b) Validation of allele types by restriction enzyme (Dde I) digestion of PCR products. M: 100 bp DNA ladder.
Figure 7. Genotyping of qSV3-2 (SNP) using a dCAPS marker. (a) Electrophoresis of PCR products from each rice genetic resource using the dCAPS marker. (b) Validation of allele types by restriction enzyme (Dde I) digestion of PCR products. M: 100 bp DNA ladder.
Agronomy 15 02050 g007
Agronomy 15 02050 g008
Figure 8. Comparison of SPAD values between SNP alleles A and G. Boxplot showing SPAD variation across the two allelic groups of the target SNP. Allele A exhibits a significantly higher median SPAD value than allele G. Outliers are shown as points above the boxplot. Asterisks (***) denote significant differences at p = 0.001 by the t-test.
Figure 8. Comparison of SPAD values between SNP alleles A and G. Boxplot showing SPAD variation across the two allelic groups of the target SNP. Allele A exhibits a significantly higher median SPAD value than allele G. Outliers are shown as points above the boxplot. Asterisks (***) denote significant differences at p = 0.001 by the t-test.
Agronomy 15 02050 g008
Table 1. Evaluation of SPAD values and coefficients of variation (CV) across 198 rice genetic resources over three years (2022–2024).
Table 1. Evaluation of SPAD values and coefficients of variation (CV) across 198 rice genetic resources over three years (2022–2024).
YearSPAD ValueHeritability (h2)Coefficient of Variation (%)
RangeMean ± SDRangeMean ± SD
202230.5~55.6(32.2~55.3)40.0 ± 5.3(40.9 ± 4.4)0.9410.2~12.85.6 ± 2.4
202334.4~58.542.8 ± 4.6
202428.0~54.040.0 ± 4.0
Heritability (h2): broad-sense heritability, calculated as the ratio of genotypic variance (σ2g) to phenotypic variance (σ2ph).
Table 2. Genome-wide association study (GWAS) results for SPAD values in 175 rice genetic resources using five association mapping models (MLM, MLMM, CMLM, FarmCPU, and BLINK).
Table 2. Genome-wide association study (GWAS) results for SPAD values in 175 rice genetic resources using five association mapping models (MLM, MLMM, CMLM, FarmCPU, and BLINK).
SNPChromosomeAlleleFrequencyPosition
(IRGSP-1.0)		−log10(p)PVE (%)MAFEffect SizeAdjusted R2Model
AX-1543706922C/T0.92/0.0834,663,7605.480.000.0911.2670.029FarmCPU
AX-1549734813G/A0.62/0.3823,418,6866.1151.240.38310.4430.077MLMM
AX-1157426723T/C0.38/0.6228,316,1549.448.190.3771.3500.056BLINK
AX-1157458373G/A0.95/0.0528,612,7037.3635.510.054−1.6350.267BLINK
AX-959557566G/A0.85/0.158,131,2805.710.000.154−2.188 ± 0.4440.169CMLM
5.710.00 −2.188 ± 0.444 MLM
AX-1539658066A/T0.40/0.608,317,40218.199.770.4002.0930.067FarmCPU
AX-1158055326T/C0.58/0.4210,023,9538.8720.480.417−2.702 ± 0.4210.241CMLM
8.8720.48 −2.702 ± 0.421 MLM
11.126.58 −2.686 MLMM
AX-1158139276T/C0.91/0.0921,892,5775.890.000.0893.5040.033MLMM
AX-1547230437T/G0.87/0.131,120,8085.742.200.137−1.4780.005BLINK
AX-2757623327A/G0.64/0.3610,746,1368.7031.770.3542.9360.068BLINK
13.1431.86 3.242 FarmCPU
AX-1157586907A/G0.61/0.3926,384,2226.122.860.3976.4780.102MLMM
AX-2817784078C/T0.91/0.091,936,3275.840.790.1001.2900.039FarmCPU
AX-15406208910T/C0.51/0.194,279,5975.650.000.486−3.508 ± 0.8330.018MLMM
Frequency: Allele frequency of the reference and alternative alleles observed among the 175 rice accessions. Position (bp, IRGSP-1.0): genomic position of each SNP in base pairs (bp) according to the International Rice Genome Sequencing Project (IRGSP) Build 1.0 reference genome. PVE: phenotypic variance explained; MAF: minor allele frequency.
Table 3. Identified QTLs associated with SPAD values detected by five association mapping models in the genome-wide association study (GWAS).
Table 3. Identified QTLs associated with SPAD values detected by five association mapping models in the genome-wide association study (GWAS).
QTLSNPAlleleChromosomePosition (IRGSP-1.0)−log10(p)Range of PVE (%)
qSV3-1AX-154973481G/A323,418,6866.1151.24
qSV3-2AX-115745837G/A328,612,7037.3635.51
qSV6AX-115805532T/C610,023,9538.87~11.126.58~20.48
qSV7AX-275762332A/G710,746,1368.70~13.1431.77~31.86
PVE: phenotypic variance explained. Position (bp, IRGSP-1.0): genomic position of each SNP in base pairs (bp) according to the International Rice Genome Sequencing Project (IRGSP) Build 1.0 reference genome.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Baek, D.-H.; Kim, T.-H.; Lee, C.-J.; Gao, J.; Park, W.-G.; Kim, S.-M. Identification of Candidate Genes Related to SPAD Value Using Multi-Year Phenotypic Data in Rice Germplasms by Genome-Wide Association Study (GWAS). Agronomy 2025, 15, 2050. https://doi.org/10.3390/agronomy15092050

AMA Style

Baek D-H, Kim T-H, Lee C-J, Gao J, Park W-G, Kim S-M. Identification of Candidate Genes Related to SPAD Value Using Multi-Year Phenotypic Data in Rice Germplasms by Genome-Wide Association Study (GWAS). Agronomy. 2025; 15(9):2050. https://doi.org/10.3390/agronomy15092050

Chicago/Turabian Style

Baek, Dong-Hyun, Tae-Heon Kim, Chang-Ju Lee, Jingli Gao, Woo-Geun Park, and Suk-Man Kim. 2025. "Identification of Candidate Genes Related to SPAD Value Using Multi-Year Phenotypic Data in Rice Germplasms by Genome-Wide Association Study (GWAS)" Agronomy 15, no. 9: 2050. https://doi.org/10.3390/agronomy15092050

APA Style

Baek, D.-H., Kim, T.-H., Lee, C.-J., Gao, J., Park, W.-G., & Kim, S.-M. (2025). Identification of Candidate Genes Related to SPAD Value Using Multi-Year Phenotypic Data in Rice Germplasms by Genome-Wide Association Study (GWAS). Agronomy, 15(9), 2050. https://doi.org/10.3390/agronomy15092050

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Article metric data becomes available approximately 24 hours after publication online.
Back to TopTop