Over the last few decades, tremendous progress has been made in increasing food output and affordability for resource-poor populations. Milled rice is composed of starch, that is, a complex carbohydrate. Rice has a slightly lower protein content than wheat, maize, and pulses, with protein being the second most essential component of cereal crops, even though little effort has been made to increase [13]. Therefore, fortification of rice with antioxidants, vitamins, modified starch, and dietary fibers are desirable characteristics to make rice a better complete staple meal at a reasonable price [8]. Genetic enhancement of such characteristics needs a thorough study of the genetic control of the trait, the genetics and molecular pathways underlying trait regulation, as well as environmental influences. The majority of these characteristics are complicated and are regulated by a large number of moderate-impact genes. The self-pollinated nature of rice enables the establishment of a variety of mapping populations, which includes F2 population, doubled haploid (DH), backcross inbred lines (BILs) and recombinant inbred line (RIL) [14]. Some newly formed innovative mapping populations such as NAM (nested association mapping) [15] and MAGIC (multiparent advanced generation intercross) populations are used to map complex traits [16].

Mapping of quantitative trait loci was investigated to determine the genetic region controlling rice nutritional quality characteristics. The rice mutants with high Fe and Zn concentrations showed Zn concentrations ranging from 15.36 to 28.95 mg/kg and Fe concentrations ranging from 0.91 to 28.10 mg/kg [17]. The complexities of nutrition quality-related characteristics vary significantly; for example, certain variables, such as folate content, have a limited number of significant QTLs, while others, such as protein content, have a large number of minor-effect QTLs. Considerable effort is being made to identify quantitative trait loci for the protein content of rice grains that are mainly located on rice chromosome segments 3 and 5 [18]. Numerous characteristics associated with nutritional quality are linked, and their QTLs typically co-localize. For example, on rice chromosome 6, retrogradation, peak viscosity, QTLs for gel consistency, amylose concentration, breakdown viscosity, final viscosity, setback viscosity and trough viscosity were identified [19]. Consequently, many improved nutritional traits are adversely controlled, making simultaneous improvement difficult. For example, grain iron content is inversely linked to grain production per plant. Similarly, the phytic acid concentration of rice affects mineral bioavailability [17].

Quantitative trait loci (QTLs) for Fe- and Zn-related characteristics from interspecific and intraspecific crosses have been documented in rice grain and less utilized in molecular breeding for this trait [20,21]. Bi-parental mapping populations are time consuming, expensive and yield a lower quality than association mapping [22]. Association mapping relies on linkage disequilibrium or differences seen in wild or cultivated species and confer a relationship between molecular marker and grain Fe and Zn contents in rice grain and heterogeneity for both traits [23]. QTL and association mapping can be utilized in rice for a variety of traits such as grain yield and attributes, seedling low temperature tolerance, cold tolerance at booting stages, heat-stress tolerance, grain quality-related traits, salinity tolerance, drought tolerance, seedling vigor and grain protein constituents [24].

The increase in QTL span and uncertainty in localization that occurs in the complication of QTL mapping applicability to the breeding program. Consensus QTLs are selected for meta-QTL analysis, and a couple of previously completed research studies are used to improve the locations of the aforementioned QTLs. Additionally, MetaQTLs are specified at the 95% confidence level [25]. In this perspective, the MetaQTL method offers an ideal chance to combine published QTL mapping information from several studies to determine more exact statistically significant levels and phenotypic changes in rice, as well as accurately characterize the QTL span. On chromosome five, one such example involves a shared QTL for phosphate and phytate [26]. Further study attempts found three MetaQTLs associated with higher Zn and Fe concentrations in rice. A similar MetaQTL study was performed to discover potential genes for salt tolerance, rice root shape, and grain size [27]. Although this research may have a synergistic as well as antagonistic impact of multiple QTLs in enhancement of nutritional characteristics, further studies are direly necessary to explore this aspect in an accurate way [17]. There are a number of genes that regulate nutritious functionality features that have been the focus of substantial research in recent years. Genes are being efficiently explored with molecular breeding, transgenic method, and even comparatively new technology like genome-editing

While the effectiveness in identifying segments of chromosome linked to characteristic QTL mapping has two significant downfalls, the QTL mapping resolution is very limited and is only used to study segregated alleles from the parent line [40]. These drawbacks of QTL mapping are eliminated by using the GWAS techniques [41]. GWAS is a technique for rapidly scanning markers throughout the whole set of DNA to identify genetic changes linked to a certain trait of several species. Following the finding of novel genetic interactions, efficient breeding methods can be used to enhance the yields of rice and other crops [42]. Additionally, the GWAS technique has numerous drawbacks, including the increase in genotype markers, diverse resources of a large germplasm collection, and allele data, such as the presence of minor alleles in at least 5% of the germplasm pool [41]. For unique alleles found in a small number of genotypes, QTL mapping is generally the best method. Recognizing the limits of both methods, it is recommended that they be used in conjunction to identify QTL [43]. The rice seed-related characteristics are analyzed by the usage of QTL mapping and GWAS analysis in combination. The concentrations of Zn, Mo, As, and Cu in 300 brown rice varieties were determined by using GWAS mapping [44]. These elements show variations in grain composition that are linked to the number of candidate genes and SNPs, and the main reason for variation is environmental circumstances [45]. The mappings of GWAS and QTL are performed in combination to investigate traits such as Al accumulates, although combined research on grain nutrient content is rare [46]. In comparison to QTL mapping, fewer efforts are undertaken to characterize nutritional quality traits using GWAS.

Molecular marker-assisted breeding is an effective strategy for incorporating desirable characteristics from a pool of high-yielding cultivars, which can only be performed with previous information on specific gene loci, associated markers, and repeated backcrossing of large segregating progenies [13]. Additionally, the recently introduced trait might not always improve as predicted, because it belongs to a diverse genetic background, and the undesired attachment leads to significant issues with marker-supported breeding [47]. [48] proposed genomic selection (GS) to overcome these constraints by estimating the potential of breeding lines of rice that are based on high-density markers and phenotypic values. Genome selection is a genetic analysis that is performed by using marker selection, in which the genetic markers of whole genome are applied to ensure the linkage of QTL with at least one marker [49]. Genome selection is being reconsidered in light of current genotyping technologies such as genotyping of the next generation. The efficacy of genome selection analysis is enhanced and made cost effective by innovative genotyping methods [50]. Despite the availability of several genotyping technologies and whole-genome-sequenced genotypes, the genome selection method takes relatively more effort for rice [51]. Genome selection is more likely to be utilized in the addition of NGS (next-generation selection) genotyping technologies in many breeding processes. The GS genotyping technique is cost effective and it increases the efficacy of genome selection technology many times. There are many genotyping methods that are publicly introduced, such as whole-genome sequences, but the usage of the genome selection method for rice genotyping is performed with minimum effort [52]. The efficacy of genome selection was studied in rice for the first time by using inbred lines to improve grain or seed quality characteristics such as height of the plant, total yield, grain yield, and blooming duration [51]. It was discovered through the combination of GS and GWAS that genomic forecast models outperformed pedigree-based prediction in predicting the phenotype [53]. This study shows that the expense of genotyping technology has increased the value of GS, and when coupled with GS and GWAS data on genetic layout and population size, rice breeding efficiency is also boosted [54]. In comparison to yield-related features, the majority of quality-related traits may be predicted accurately. Because quality characteristics have a greater heritability, implementing GS becomes easier [50].

Many other research findings on the assessment of colored rice for various vitamins, antioxidant compounds and minerals have noticed considerable variation, and these accessions were found to contain three to four times more nutrients than advanced rice varieties [55]. In a research study, 30 (53%) quantitative trait loci were co-located with identified or functionally related genes. OsZFP252, OsHMA9, OsNRAMP7, OsMAPK6, and OsMADS13 were among the significant candidate genes for grain Zinc (Zn). Sayllebon, a red rice genotype that is high in both anthocyanins and zinc, could be a valuable breeding material for nutritious rice. QTLs may be utilized for both QTL pyramiding and genomic selection. Some of the discovered QTLs may be validated further by detailed mapping and functional characterization [56].

A genome-wide association study (GWAS) was conducted, and 29 marker–trait associations (MTAs) with significant relationships for characteristics, including ZnMR (5 MTAs), FeBR (6 MTAs), FeMR (7 MTAs) and ZnBR (11 MTAs) [57]. The co-localization of the MTAs controlling the linked features indicates the prospect of their improvement throughout. The associated robust MTAs could be a valuable source of information for enhancing Fe and Zn concentration in rice grain and addressing Fe and Zn malnutrition among rice consumers [58].

The GEBV (genomic estimated breeding values) computed using the GS technique demonstrated a broad range of reliability for characteristics within the rice plant such as flowering time, plant height, grain yield, and panicle weight. To understand the impact of population structure and marker density on the reliability of genomic prediction, researchers may also look at the structure of characteristics, as well as the reliability of prediction based on genotype [59]. In 2014, a novel approach termed genomic hybrid breeding for the prediction model was suggested with the combination of epistasis and dominance [60], it being a combination of phonological modeling and genome prediction to enhance the phenotypic prediction of complex traits among various settings for genomic hybrid breeding of rice [61]. While genome selection is increasingly utilized to examine rice quality features, the investigation into its efficacy in evaluating nutritional aspects remains lagged.