An overview of genetic mapping in plants
The question of which genes contribute to which phenotypic characteristics/traits have been well addressed by an approach known as genetic mapping. In 1911, T.H. Morgan and his student Alfred Sturtevant, published the first genetic map, depicting the location of six sex-linked genes on the chromosome of a fruit fly. In plants, genetic mapping serves as the foundation for the identification of genomic regions underlying important plant traits such as disease resistance, salt, and drought tolerance/resistance, yield and quality characteristics (Sharlach et al., 2013; Long et al., 2013; Solis et al., 2018; Pereira et al., 2018; Barik et al., 2020). These are complex traits controlled by polygenes or minor genes and hold great economical and agricultural importance in the era of climate change, malnutrition and food security. By far polygenic traits have been proved to be challenging due to complex segregating patterns and complications in the precision mapping of all genomic regions responsible for the variation in complex traits (Doerge 2002). The evolution of species occurred in complex environments with varying conditions, both in time and space. These complex traits can also be designated as quantitative traits and are strongly influenced by fluctuating environments; for instance, heritability measured under controlled conditions (laboratory or greenhouse) often exceeds field experimentation (Anderson et al. 2014). A genetic locus/loci or more precisely a gene linked to the natural variation in quantitative trait/s is known as quantitative trait locus/loci (QTL). The hypothesis that phenotypic variation of complex traits may be linked to several structural variants or causative variants, catalyzed the genetic mapping studies in plants (Xu et al. 2017).
The identification of QTL linked to natural phenotypic variation is the main aim of genetic mapping (Xu et al. 2017). It is also referred to as meiotic mapping or linkage mapping and utilizes molecular markers in various plant species to reveal causative genomic regions, intended to determine the relative position of genes on a molecule of DNA (plasmid or chromosome), along with the distance between them. A genetic map showed a position and relative genetic distance between the markers along chromosomes, which is analogous to a flanking region cuddled to the gene of interest (Semagn et al. 2006). In fact, a genetic map is synonymous with a highway mile-marker system created by geneticists. The highway mile marker system was created through the development of structured plant populations which showed segregation for the set of markers along with the trait of interest. The distance between the markers residing on the same chromosomes can be determined by detecting the amount of recombination in the structured plant populations. Recombination is a vital meiotic process occurring in the anther sacs and ovule of plant species involving the exchange and repair of male and female chromosomes. Higher recombination indicated that the markers are farther apart from each other (Hyten and Lee, 2016).
Numerous markers like protein markers, phenotypic markers and molecular markers are used for the construction of genetic linkage maps. The advancements in next-generation sequencing (NGS) have made available high-throughput markers for fine mapping of plant genomes at a reasonable cost. In this regard, the most common sequence variations or molecular markers that can serve as potential genetic markers include simple sequence repeats (SSRs) and single nucleotide polymorphisms (SNPs). Recently, genetic linkage mapping has expanded and got several applications such as QTL mapping, gene mapping, genome-wide association mapping (GWAS), and marker-assisted selection (MAS) (Hyten and Lee, 2016). With the availability of NGS, linkage mapping can be done to locate QTL (Quantitative trait loci) on corresponding genomes of the biparental population to facilitate MAS. Alternatively, QTLs or genomic regions can be identified by extensive genotyping of germplasm panel of breeding populations or cultivars depicting biological variation/diversity in phenotypes across fluctuating environments, with an objective to reveal marker-trait associations. The former approach is termed QTL mapping while the latter is known as genome-wide association mapping (GWAS).
Association Mapping in Wheat
According to FAOSTAT, 2020 the wheat crop stands at the second position among the most-producing cereal crops around the world (FAOSTAT, 2020). It contributes around 19% among the total production of cereal crops and delivers about 55 % of the carbohydrate which are utilized by humans all over the globe (Gupta et al. 1999; Bagge et al. 2007). Biotic and abiotic stresses effect the production of wheat and plays a significant role in the reduction of its potential yield. It’s the requirement of the time to make strategies and develop a technique against these stresses, and plant breeding is one of the tools to cope with biotic and abiotic stress conditions including drought, salinity and plant diseases (Araus et al. 2008; Cooper et al. 2009). Traditional plant breeding approach is very slow and still there is a big yield difference between drought prone areas and perfect growing areas for most of the crops, comprising wheat. As drought tolerance is a very complex quantitative trait that is controlled by minor genes which are located on several chromosome regions (Barnabas et al. 2008; Fleury et al. 2010; Pinto et al. 2010; Ravi et al. 2011; Mir et al. 2012). Genetic enhancement for drought resistance necessitates to find the genomic regions related to drought tolerance and transferring the genes possess the desired characters to locally adapted cultivars. The major task for applying this approach in future breeding programs is the finding of the most appropriate targeted traits in an efficient and cost-effective method for varied water scarcity scenarios (Passioura 2012). Recently improvements in high-throughput genotyping and phenotyping have increased the understanding of biological and molecular bases hold complex traits including drought resistance (Collins et al. 2008; Habash et al. 2009; Mir et al. 2012; Sinclair 2012). QTL mapping is an important tool for the better understanding of genomic structure of more complex characters in the plants (Holland 2007). Yet, QTL mapping technique by consuming biparental populations defines a minor portion of the genomic structure of a parameter for the reason that only two alleles on a locus can be screened at a time. Some other restrictions of biparental populations are low resolution of mapping, population specificity of identified QTL, and extensive time is necessary to generate the mapping populations. These restrictions make hurdles in the transfer of information from bi-parental QTL studies to the practical use in breeding programmes. Few problems belong to the bi-parental mapping population has been resolved with the introduction of the association mapping technique. Diverse germplasm can be used for association mapping and high resolution from AM can aid to detect QTL for many traits which make this technique most effectual and less costly than bi-parental QTL mapping approach (Breseghello and Sorrells 2006; Ersoz et al. 2009; Sorrells and Yu 2009; Waugh et al. 2009).
Mostly agronomic traits were organized by many genes, and QTLs trainings in these plants commonly use the linkage studies with F2, DH and backcross populations. In wheat, QTL mapping studies have been performed on agronomic traits such as flowering period, grain yield, some quality traits and resistance against diseases by using these populations (Borner et al. 2002; Peng et al. 2003; Blanco et al. 1998; Khan et al. 2000; Perretant et al. 2000; Olmos et al. 2003; Otto et al. 2002; Gervais et al. 2003; Navabi et al. 2005; Toth et al. 2003; Hanocq et al. 2004). Contrarily, QTL mapping in segregating population have a lot of limitations. This includes less allele quantity and middle density of marker. Association mapping is a technique that works on the basis of linkage disequilibrium (Zondervan and Cardon 2004). Compared to conventional linkage analysis, association mapping offers the vital benefits of sampling unrelated individuals in the population for marker-assisted selection in diversified traits study in plant breeding (Risch 2000; Jannink et al. 2001). First of all, the investigational population can be a demonstrative sample of the population to which interpretation is wanted. Secondly, association mapping could be more effective in the utilization of resources and means. For example, many traits could be studied in the same population by utilizing the similar genotypic records. A greater quantity of molecular markers with polymorphic is expected to deliver improved genomic analysis than biparental mapping. In addition, phenotypic record of many years and time would be accessible deprived of extra cost in study of elite wheat lines (Rafalski 2002). Recently, it has been extensively used for the QTLs discovering in many plants’ species like maize, rice and barley etc., Important SSR markers linked with the size of wheat kernel was determined by Breseghello and Sorrells (2006) in hexaploid wheat. International Maize and Wheat Improvement Centre (Jose et al. 2007) set up a multi-environmental experiment to examine the relation among stem rust, leaf rust, yellow rust, powdery mildew, and grain yield by utilizing two markers diversity array technology (DArT) marker and SSR marker. Tommasini et al. (2007) found genomic region that controls resistance against Stagonospora nodorum blotch by linkage analysis with SSR markers. Association mapping works on the principle of linkage disequilibrium (LD). Extent of LD across the genome determines the strength and resolution of association mapping studies. The strength of association mapping increases when linkage disequilibrium decays quickly (Rafalski 2002). Extent of LD is varied in different species. The extent of linkage disequilibrium was about 1 cM in chromosome 2D, nearly 5 cM in the centromeric region of chromosome 5A and less than 0.5 cM in chromosome 3B of hexaploid wheat depends on SSR markers (Breseghello and Sorrells 2006; Tommasini et al. 2007). Since linkage disequilibrium is affected by many dynamics such as population history, recombination rate and mating system, it is desirable to examine it in each study.
Association mapping studies in hexaploid wheat are extended to grain yield, kernel size and milling quality of grain, the amount of glutenin content and disease resistance (Breseghello and Sorrells 2006; Jose et al. 2007). On the other hand, there are limited studies about yield related agronomic trait like spikelets per spike and grains per spike. Some examples of association mapping studies in wheat are presented in Table 1.
Table 1
Examples of Association Mapping studies in Wheat
Population
|
Sample Size
|
Markers Used
|
Traits
|
References
|
Diverse Accessions
|
1596
|
5011 SNPs
|
Seedling leaf rust resistance
|
Li 2016
|
Diverse cultivars
|
95
|
95 SSRs
|
Kernal Size, Milling Quality
|
Breseghello and Sorrells 2006
|
Soft Winter Wheat Accessions
|
95
|
36 SSRs
|
Kernal Size, Milling Quality
|
Flavio et al. 2006
|
Diverse Accessions
|
108
|
85 SSRs
40 EST-SSRs
|
Agronomic Traits (Plant Height, Spike Length, Spikelet per spike, Grains per Spike, 1000 grain weight
|
Yao et al. 2009
|
Elite durum wheat accessions
|
189
|
56 SSRs
|
Drought + Adaptive traits and grain yield
|
Marco Maccaferri et al. 2011
|
Diverse European Elite wheat lines
|
207
|
115 SSRs
|
1000- Kernal weight, Protein content, Sedimentation volume, Starch concentration
|
Reif et al. 2011
|
Elite winter wheat
|
120
|
3051 DArT
|
Yield and grain quality
|
Tadesse et al., 2015
|
Durum Wheat
|
287
|
30155 SNPs
|
Agronomic traits
|
Mengistu et
al., 2016
|
Spring type bread wheat cultivars
|
108
|
9646 SNP
|
Yield and yield related traits
|
Qaseem et al., 2018
|
Spring wheat
|
586
|
90K SNP
|
Grain yield and agronomic traits
|
Garcia et al., 2019
|
Advanced lines
|
382
|
2214 SNPs
|
Drought + agronomic traits
|
Ballesta et al., 2019
|
Association mapping in Rice
Rice is a staple food crop feeding 3 billion people worldwide (FAO, 2021). The development of improved varieties providing resistance against biotic and abiotic stress factors combined with high productivity, quality and adaptability under wide range of environmental conditions are the major objectives and challenges of rice breeders worldwide (Gregoria et al., 2013). Rice genome is small having a size of 400 to 430 Mbp among major cereal crops (Eckardt, 2000). A genome size of 389 Mbp was reported by International Rice Genome Sequencing Project (IRGSP). A total of 37,544 protein coding sequences were detected along with the identification of 80,127 polymorphic sites that distinguish two rice species i.e., indica and japonica (International Rice Genome Sequencing Project, 2005). It is fortunate that the availability of sequencing technologies coupled with fully sequenced rice genome (availability of 3000 re-sequenced varieties) has created an opportunity to enhance crop productivity and quality using modern genomic tools and methods (Begum et al., 2015; McCouch et al., 2016). GWAS is a significant tool for unveiling genotypic variation associated with complex phenotypic traits. Mapping of candidate genes concerning to the desired trait of interest can be achieved efficiently through GWAS. Several methods such as GLM, MLM, FarmCPU and BLINK were employed for GWAS analysis. Though GLM helps in the identification of a greater number of QTLs compared to other methods but prone to false-positives. MLM reduce false positive rates but may causes overcorrection which reduced its power of detection. Both BLINK and FarmCPU uses multilocus model for testing markers across genome, hence possesses higher statistical power and decrease false-positives (Zhong et al., 2021). A comprehensive detail of GWAS in rice is outlined in Table 2.
Table 2
Examples of Association Mapping studies in Rice
Phenotypic Trait
|
Markers
|
Population
|
Number of associated loci/QTLs
|
Chromosomes (Position)
|
References
|
Agronomic & Yield traits
|
|
|
|
|
|
Yield and Agronomic traits
|
71,710 SNPs
|
363 elite breeding lines from IRRI
|
52
|
2, 3 (233 bp to 900 kb), 6 (47.9 kb), 7, 8, and 11 (38 bp)
|
Begum et al. (2015)
|
Agronomic traits and drought tolerance
|
3.6 million SNPs
|
517 landraces
|
80
|
2 (25.02 and 30.18 Mb), 3 (17.37 Mb), 5 (5.3 Mb), 6 (1.7-6.7 Mb), 9 (7.3 Mb), 10 (2.31 Mb) and 11 (21.16)
|
Huang et al. (2010)
|
Agronomic traits
|
426,337 SNPs and 67,544 indels.
|
176 japonica rice varieties
|
26
|
1, 3, 6, 7, 8and 11
|
Yano et al. (2016)
|
Panicle architecture
|
411,066 SNPs
|
421 accessions (Rice diversity panel)
|
106
|
All chromosomes
|
Zhong et al. (2021)
|
Morphological traits of panicle
|
241 DArT and 25971 SNPs
|
159 traditional varieties
|
105
|
All chromosomes except 5
|
Ta et al. (2018)
|
Abiotic traits
|
|
|
|
|
|
Cold tolerance
|
36727 SNPs
|
211 landraces
|
12
|
2 (4.4 Mbp); 3 (10.23-11.31 Mbp); 4 (11.4 Mbp); 7 (29.11 Mbp) and 9 (7.10 Mbp)
|
Li et al. (2021)
|
Chilling tolerance
|
148 SSRs+3 InDels+6 SNPs
|
202 RMC accessions
|
48
|
All chromosomes; two novel QTLs (qLTSS3-4 and qLTSS4-1) positioned at 35.3 & 13.6 Mbp on chr 3 & 4, repectively.
|
Schläppi et al. (2017)
|
High temperature
|
14,779,691 SNPs data from 3 k database
|
255 Asian cultivated rice varieties
|
117
|
3 (2.96-3.82 cM), 6 (2.77 cM), 7 (3.68 cM), 9 (3.54 cM), 11 (5.01-6.72 cM),12 (2.82, 4.80 cM)
|
Wei et al. (2021)
|
High temperature
|
2 million SNPs
|
98 rice accessions
|
2
|
10 and 11
|
Kwon et al. (2021)
|
Drought tolerance
|
21623 SNPs
|
180 Vietnamese landraces
|
17
|
1 (22.97-24.84 Mbp), 2 (35.13 Mbp), 5 (14.99-15.94 Mbp), 6 (17.77-17.84 Mbp), 7 (17.90-20.79 Mbp), 10 (10.28 Mbp), 11 (24.38-25.67; 66.42-68.95 Mbp)
|
Hoang et al. (2019)
|
Drought tolerance
|
170 SSR
|
114 rice genotypes
|
11
|
1, 4, 7 and 9
|
Verma and Sarma (2021)
|
Drought tolerance
|
4 358 600 SNPs (RiceVariation Map v2.0)
|
507 accessions
|
470
|
Nearly all chromosomes (OsPP15 gene)
|
Guo et al. (2018)
|
Drought tolerance
|
4,358,600 SNPs
|
529 rice accessions
|
143
|
All chromosomes (Nal1 and OsJAZ1 genes)
|
Li et al. (2017)
|
Salt tolerance
|
> 33000 SNPs
|
155 varieties
|
151
|
1 (40.79-42.98 Mbp); 2 (35.26 Mbp); 6 (29.76-31.21 Mbp); 7 (19.66-19.83 Mbp); 9 (17.96 Mbp); 10 (6.12-7.32 Mbp); 12 (17.53-18.60 Mbp)
|
Nayyeripasand et al. (2021)
|
Cadmium tolerance
|
3.3 million SNPs
|
188
|
119
|
All chromosomes
|
Yu et al. (2021)
|
Sulfur deficiency
|
700k SNPs
|
98 accessions
|
11 (low S conditions)
|
1 (12.11 Mb), 2 (21.4 Mb), 3 (19.34, 28.01 Mb), 4 (32.96 Mb), 6 (0.29 Mb), 9 (15.01 Mb) and 11 (17.95 Mb)
|
Pariasca-Tanaka et al. (2020)
|
Biotic traits
|
|
|
|
|
|
Rice Blast resistance (Maganaporthe oryzae)
|
High density 700 K SNP array
|
420 accessions
|
97
|
Distributed on all chromosomes
|
Kang et al. (2016)
|
Rice blast (Magnaporthe oryzae)
|
High density 700 K SNP array
|
413 accessions
|
16
|
1, 3, 4, 5, 8, 9, 10, 11, 12
|
Zhu et al. (2016)
|
Sheath Blight resistance
(Rhizoctonia solani)
|
44k SNPs
|
299 varieties
|
147
(2 reliable QTLs)
|
3 (16-17 Mbp); 6 (20-22.5 Mbp)
|
Chen et al. (2019)
|
Rice black-streaked
dwarf virus
|
44k SNP array
|
305 accessions
|
13
|
1 (2.09-2.72 Mbp); 2 (19.54-29.73 Mbp); 3 (29.7 and 27.69-28.02 Mbp); 4 (41.72-52.55 Mbp); 6 (10.95-11.10 and 17.99-19.86 Mbp); 8 (16.3 and 19.7 Mbp); 11 (9 Mbp)
|
Feng et al. (2019)
|
Bacterial leaf streak resistance (Xanthomonas oryzae)
|
140345 SNPs
|
510 accessions
|
79
|
All chromosomes (majority on chr 11)
|
song et al. (2021)
|
Quality traits
|
|
|
|
|
|
Grain protein, amylose content, alkali spreading value
|
22947 SNPs
|
217 USDA accessions
|
10
|
4 (23.52 Mbp); 6 (16.43-17.88 Mbp); 8 (18.11 Mbp); 12 (24.08 Mbp)
|
Song et al. (2019)
|
Grain appearance and milling quality
|
22488 SNPs
|
258 accessions
|
72
|
All chromosomes
|
Wang et al. (2017)
|
Cooked rice texture
|
700K high Density Rice Array (HDRA)
|
236 Indica accessions
|
97
|
All chromosomes
|
Misra et al. (2018)
|
Grain Starch
|
246,026 SNPs
|
115 accessions
|
16
|
1, 2, 3, 5, 6, 8, 11
|
Biselli et al. (2019)
|
High density panel of markers covering the whole genome can be used successfully for the detection of recombination density points in unrelated populations (Zhu et al., 2008). High density polymorphic 71,710 SNPs were genotyped using genotyping by sequencing (GBS) technique and further utilized to perform GWAS in 363 elite panel of breeding lines from International Rice Research Institute (IRRI). 52 QTLs were identified and found to be associated with yield and yield components such as flowering time, grain length, width, plant height and rice grain yield. Plant height and flowering time were linked with the candidate gene named OsMADS50 positioned on chr 3, which is a known activator of flowering in rice and also termed as pleiotropic gene (Lee et al., 2004; Begum et al., 2015). A high-density haplotype map was constructed using approximately 3.6 million SNPs through a sequencing of 517 rice landraces (Oryza sativa indica). The information was further used to perform GWAS regarding 14 agronomic traits. 56 significant SNPs were detected that are tightly linked to already identified genes in rice on chr 3, 5, 6 and 7. The peak signals were identified for apiculus colour, pericarp colour, Gelatinization temperature, amylose content, grain width and grain length (Huang et al., 2010). 29 stable QTLs were identified on nearly all 12 chromosomes except 5 using 241 DArT and 25971 SNPs in a set of 159 traditional rice varieties. These QTLs were associated with genes that control panicle development and architecture such as LONELY GUY (LOG), TAWAWA1 (TAW1) and RICE DOF DAILY FLUCTUATIONS 1 (RDD1) (Ta et al., 2018). It is common to select alleles of the same gene during crop domestication and breeding resulting in allelic heterogeneity. To deal with spurious associations, gene-based association analysis proves to effective. In rice, chromosome 8 contains a crucial casual gene (LOC_Os08g37890) for awnless phenotype detected through gene-based association analysis facilitated by whole genome sequencing. This method followed by gene expression profiling eliminates spurious associations that are meant to happen through allelic heterogeneity (Yano et al., 2016). It is pertinent to discuss here the power of GWAS combined with GPWAS (Genome-Phenome wide association mapping) to unveil significant QTLs and genes linked with panicle architecture in rice. 23 candidate genes were identified by employing both these techniques using 411,066 SNPs covering 62.40% (23,623 out of 37,860) genes in rice (Zhong et al., 2021). A candidate gene (Os07g0669700) on chr 7 with gene annotation of potassium transporter 7 (OsHAK7) was associated with total spikelet number per plant (TSNP). Loss of function OsHAK7 mutant causes a decrease in panicle length, grain yield and seed setting in comparison to wild type (Chen et al., 2015). Os01g0140100 was TSNP linked overlapped gene identified by both GWAS and GPWAS that is crucial for rice fertility. Potassium transporter gene (Os02g0809800; OsPHO1;2) detected by GPWAS technique was known to regulate number of panicles, grain number per panicle, 1000-grain weight and yield per plant in rice (Zhong et al., 2021).
A study conducted by Schläppi et al. (2017) concluded that japonica rice accessions possess a high tendency to tolerate chilling environment as compared to indica accessions. Mapping of QTLs using GWAS approach was done based on five chilling tolerance indices (namely low temperature germinability (LTG), plumule growth rate after cold germination (PGCG), seedling survivability at low temperature (LTSS), plumule recovery growth after cold exposure (PGC) and survival at low temperature (LTS)). 48 QTLs were identified on nearly all 12 chromosomes. qLTG10-2 located on chr 10 (21.1 Mbp) confer chilling tolerance due to the overexpression of associated candidate gene MYBS3 in rice (Su et al., 2010). Similarly, qPCGC9-2 is present within the candidate gene OsWRKY76, which showed tolerance against cold during overexpression studies (Yokotani et al., 2013). Schlappi et al. (2017) identified two novel QTLs (qLTSS4-1 and qLTSS3-4) associated with seedling chilling tolerance, which provides a good indicator of field performance of transplanted rice seedlings. An important QTL linked with recovery of plumule after exposure to cold stress was found on chromosome 6 (qPGC6-1). This particular QTL was found to be associated with cold stress gene OsDREB1C. It is pertinent to recall here that DREB transcription factor family is highly responsive in improving tolerance to drought and low temperature conditions (chawade et al., 2013). Chilling acclimation has been achieved through genes such as OsCAF1B and ABA receptor OsPYL/RCAR5. These annotated genes were associated with QTLs located on chr 3 (qLTG3c), 4 (qLTG4), 5 (qLVG5), 7 (qLVG7-1) and 12 (qCTGERM12-2). Overexpression of these genes enhanced survival rate and decrease the leakage of electrolytes in transgenic lines subjected to cold stress (Verma et al., 2019; Fang et al., 2021). Population structure analysis conducted by Li et al. (2021) also confirmed that japonica rice was more tolerant against cold as compared to indica rice in a GWAS study carried for 211 rice landraces. Using MLM with a robust set of 36,727 SNPs, twelve QTLs were associated with cold tolerance at bud burst stage (seedling survival rate). Seven novel QTLs were identified and among those qSSR9 was the most significant SNP explaining the maximum phenotypic variation (R2= 0.14). This QTL was localized on chr 9 comprising of 39 underlying candidate genes. Further, candidate gene analysis elucidates that among 39 genes, 5 were colocalized with cold tolerance genes such as COLD1, LTG1, Ctb1, OsbZIP73 and OsWRKY71, respectively. Therefore, the region located on chr 9 is favourable for mining of alleles and genetic resources to be used for breeding new rice varieties characterized with cold tolerance (Li et al., 2021). In rice, chromosome 9 contains a QTL named qSR_ind9–3 associated with a gene Ugp1 that confers tolerance to heat stress at the seedling and heading stage. GWAS revealed two more significant QTLs i.e., qDW_ind8 (3.09 Mb–3.29 Mb) and qFW_ind8 (3.09 Mb–3.29 Mb) positioned on chromosome 8. This region was linked to dry and fresh weight of rice controlling biomass under high temperature stress environments. Interestingly, in rice the LD is between 100 to 200 kb. A 200 kb interval in qFW_ind6, qFW6 (chr 6), and qDW_ind7, qDW7 (chr 7) carries important candidate genes conferring resistant to heat at the seedling stage (Wei et al., 2021). Functional genomic study of these genes will further disclose their molecular mechanisms. Kwon et al. (2021) were of the view that chromosomes 10 and 11 bear two significant QTLs (qHTS10 and qHTS11) associated with spikelet fertility under high temperature (marker for tolerance to high temperature) revealed in a set of 98 rice accessions using GWAS technique. Haplotype analysis showed 39 candidate genes located within these two QTLs. Interestingly only one gene (Os10g0177200: EF-HAND 2 domain containing protein) demonstrated significant difference among the haplotypes. This candidate gene encodes Ca+2 binding proteins (CBP) having a helix-loop-helix structure conferring resistance against stress environments (cold and heat) through the movement of Ca+2 ions within the cell (Zeng et al., 2017). RM212 and RM252 markers identified through association mapping technique have been linked with drought tolerant traits in rice (Verma and Sarma, 2021). QTLs such as qDTY1.1, qDTY2.2, qDTY2.3, qDTY3.1, qDTY3.2, qDTY6.1, and qDTY12.1 regulates grain yield under drought stress (Li et al., 2017; Hoang et al., 2019). Candidate gene analysis demonstrates the efficiency of two genes (Nal1 and OsJAZ1) controlling root traits improving drought avoidance in rice (Li et al., 2017).
GWAS identified 11 SNPs associated with sheath blight (SB) resistance in rice diversity panel of 299 varieties using 44k SNPs. Two reliable QTLs (qSB3 and qSB6) localized on chr 3 and chr 6 were found to be linked with SB resistance (Chen et al., 2019). Using 700k SNPs in an unrelated population of 420 rice accessions, GWAS identified 97 non-redundant loci associated with blast resistance (Magnaporthe oryzae). Chromosome 9 contains LABR_64 locus associated with broad-spectrum blast resistance against all five tested blast isolates in several different subpopulations of rice (Kang et al., 2016). This locus colocalizes with previously known resistance locus Pi5 that requires two NBS-LRR (nucleotide-binding site leucine-rich repeat) genes i.e., Pia and Pikm, also found in R genes (1). It might explain the broad-spectrum biotic resistance of this particular locus. Bacterial leaf streak (BLS) is a devastating rice disease caused by Xanthomonas oryzae pv. Oryzicola. GWAS performed on 510 rice accessions revealed 69 QTLs associated with BLS with a peak signal on chromosome 11 followed by chr 1 and chr 5. Important QTLs includes qnBLS5.1, qnBLS9.1 and qnBLS11.17 corresponding to genes LOC_Os11g47600 and LOC_Os11g47680 which perform functional roles under stress conditions (Xie et al., 2021). Similarly, GWAS identified important QTLs associated with quality characteristics of rice as outlined in Table 2. Significant marker trait associations were detected on chr 1, 5 and 6 related to starch and amylose contents (Biselli et al., 2019). A major QTL corresponding to rice textural attributes such as adhesiveness and amylose content were positioned on chromosome 6 (Misra et al., 2018). A total of 19 candidate genes including GS3 and TUD and 17 QTLs including two previously mapped QTLs (qGRL7.1 and qPGWC7) were discovered by Wang et al. (2017) affecting rice milling properties and grain appearance. GWAS approach played a crucial role in development of tolerant/resistance rice varieties and paved a way for next generation breeding using selection markers. In addition to marker assisted based breeding, candidate genes provide valuable information for functional characterization and development of molecular tool kit for continued rice improvement.