Genetic variability study in Bread Wheat ( Triticum Aestivum L . ) under Temperate Conditions

Wheat is one of the most important crop among the prime cereals at the global level. In the present investigation, twenty-four bread wheat (Triticum aestivum L.) genotypes were evaluated to estimate the extent of genetic diversity. Analysis of variance revealed that genotypes possess significant genetic variability among all traits at p < 00.1. It was also observed that HPW 447 had recorded the highest value of 35.16 (q/ha) whereas, HPW446 exhibited the lowest value (23.78 q/ha) for the yield. A significant positive correlation was observed among yield and biological yield traits. On contrary, yield and days to maturity revealed a negative correlation. Days to heading revealed a significant positive correlation with plant height (0.40) and a negative correlation with biological yield (-0.45), respectively. Diversity analysis classified the evaluated wheat genotypes into four (4) distinct groups based on an index of similarity and dissimilarity of attributing traits. Group I and II have one genotype each viz., HS637 and HPW447, respectively whereas in third group 6 genotypes including HPW441, HPW444, HPW446, VL2030, HS634, and VL2025 are clustered together. The fourth group had two sub-groups. The first sub-group had five (5) genotypes and the second sub-group had eight (8) genotypes. The first three principal components based on the Euclidean similarity matrix explained 45.13, 17.85 and 14.71 percent of the total variation, respectively. It is concluded that these wheat genotypes (HS631, HPW 442, HPW 447, VL 2027 and VL2028) possess significant genetic variability and requisite potential for use in breeding as candidate wheat genotypes under temperate conditions. It is further suggested that by multivariate methods diverse parents can be identified with a scope in generating transgressive segregants for prospective breeding strategies in the improvement of the wheat crop. Current Agriculture Research Journal Journal Website: www.agriculturejournal.org ISSN: 2347-4688, Vol. 6, No.(3) 2018, pg. 268-277 CONTACT Shabir H. Wani shabirhussainwani@gmail.com Mountain Research Centre for Field Crops, Khudwani Anantnag -192101, Sher-e-Kashmir University of Agricultural Sciences and Technology of Kashmir, J & K, India. © 2018 The Author(s). Published by Enviro Research Publishers. This is an Open Access article licensed under a Creative Commons license: Attribution 4.0 International (CC-BY). Doi: http://dx.doi.org/10.12944/CARJ.6.3.06 Article History Received: 7 August 2018 Accepted: 1 December 2018


Introduction
Wheat (Triticum spp.) is the second important food crop after rice worldwide 1,2 .It provides 20% of the calories and protein for the global population 3 .Plant genetic diversity decides their prospective competence and uses for breeding, to enhance and improve food production 4 .The existence of genetic diversity and variability play a crucial role in formulating a tangible and successful breeding programme.Genetic diversity is indispensable to convene the diversified target of plant breeding such as breeding for increasing yield, wider adaptation, desirable quality, pest and disease resistance 5 .At various international research centers around the globe wheat breeders are working for the improved grain yield with better quality along with important agronomic traits, hence the information of the genetic diversity within a germplasm collection has a significant impact for crop improvement and vital for development of wheat varieties resilient to climate change and diverse environment.Hybridization and subsequent selection are one of the most successful approaches utilized for wheat breeding 6 .The baseline strategy in any breeding programme depends primarily on the target trait and subsequent choice of parents to be involved under hybridization programme.Transgressive segregation can be helpful when parents used in hybridization are genetically dissimilar 7 .The genetic distance between parents is directly proportional to the extent of heterosis found in progenies 8 .Experiments on genetic variability provide a base of information regarding trait wise variation in the experimental material.Correlation studies helps to quantify and evaluate the proportion of the phenotypic correlation associated with genetic backgrounds, to investigate whether the selection for a particular trait affects more traits, to examine indirect gains due to selection on correlated traits, and to dissect the complexity of the traits.Therefore, keeping in mind the above facts, we investigated the extent of genetic variability present in a set of bread wheat genotypes for various traits and the phenotypic correlation coefficients between yield and component traits.

Materials and Methods
Twenty four bread wheat (Triticum aestivum L.) genotypes from the Indian Institute of Wheat and Barley Research (IIWBR), Karnal under the All Indian Coordinated Wheat Improvement Project were used in this investigation (Table 1).
These were sown and evaluated under rainfed condition during rabi season of 2016-17, in a randomized complete block design with three replications with the plot size of 3.5x1.2m at the The geophysical coordinates of the location fall between 34° N latitude and 74° E longitude at an altitude of 1560 m above mean sea level.The observations were recorded on diverse morpho-physiological and yield attributing traits, viz., Days to heading, days to maturity, plant height (cm), 1000 grain weight (g), biological yield (Kg/plot) and yield (Q/Ha) per hectare for conducting genetic diversity analysis.Recommended package of practices was followed to raise the crop.Analysis of variance (ANOVA) and mean comparison were performed using SAS ver.9.1 9 .Descriptive statistics and correlation analysis were performed Using SPSS Ver.23 10 .Cluster analysis and PCA was performed using R 11 , PAST 12 and STAT graphics Ver.XVII 13 .

Analysis of variance and mean comparison
The analysis of variance revealed (Table 2) significant difference in all the investigated traits.The phenotypic values for days to heading ranged from (175-190) days as observed in case of wheat genotypes HPW446 and VL2025 (190 days) and HPW447 (175 days), respectively (Table 3 and Table 4).The experimental set exhibited a significant genetic variability for days to maturity as wheat genotypes VL2026 and HPW441 matured in 226 days whereas, VL2029 matured in 219 days (Table 3 and Table 4).HS635 was the tallest (130 cm) and HPW447 and VL2025 were the shortest (93 cm) genotype with respect to plant height.For the trait, 1000 grain weight, UP2991 and UP2990 had the highest value (42 g) and VL907 and HPW447 had the lowest value (37.5 g).It was also observed that HS631 and HS633 recorded the highest value (4.7 kg/plot) and HPW446 had the lowest value (3.02 kg/ plot) for the biological yield (Table 3 and Table 4).

Correlation Analysis
Phenotypic correlation analysis was conducted between different morpho-physiological traits and it was observed that only a few of them exhibited a significant correlation coefficient at 0.01 or 0.05 level of significance (Table 5).It is evident that there is a significant positive correlation for days to heading with plant height (0.40) and a negative correlation between biological yield (-0.45) respectively.It has also been observed that biological yield (kg /plot) and an overall yield are highly correlated (0.619) with each other.The other parameters do not depict any significant correlation with yield traits.

Cluster Analysis
Cluster analysis was conducted to assess the quantum of genetic variability within and between the discrete groups based on the index of similarity and dissimilarity as indicated by the genetic distance between them.Distance coefficient between individuals was calculated using the Euclidean square distance method, and different cluster analysis methods like Ward, Nearest, and Fartherest neighbors.Ward method was more efficient and Fig. 2: PCA for the quantitative traits (yield-related traits) effective in grouping the genotypes that were further cross-validated by using discriminate analysis (Figure 1).Based on the dendrogram (Figure 1), it can be clearly expressed that the wheat genotypes are classified into 4 major groups with distinct phenotypic features.Group I and II are solitarily represented by one distinct genotypes, viz., HS637 and HPW447, respectively.In the third group of wheat genotypes viz.HPW441, HPW444, HPW446, VL2030, HS634, and VL2025 are clustered together based on the minimums classifier value for biological yield.In the fourth group, genotypes have been clustered into two sub-groups.The first sub-group had five (5) and the second had eight (8) genotypes, respectively.These genotypes exhibited highly desirable and significant genetic diversity with respect to diverse morphological and yield attributing traits.For example genotype, HPW 447 dwarf, early maturing and high yielding.Maximum distance was observed between genotypes HS631 and HS637.Minimum distance was observed among genotypes HS631 and UPL2990, respectively.

Principal Component Analysis (PCA)
Figure 2 represents the PCA analysis of standardized log transformed and significantly correlated (at 0.001 level) quantitative phenotypic traits for 24 wheat genotypes.It was observed that the first three main Eigenvalues on the Euclidean similarity matrix explained 45.13, 17.85 and 14.71 percent of the total variation, respectively with a cumulative variance of 77.69 of total phenotypic variation.It is revealed (Table 6) that days to heading, biological yield, and yield/ha contribute significantly to the quantum of  3).
It is evident from Figure 2, that vectors pertaining to days to heading, biological yield and yield/ha are on the opposite direction that indicates the relative importance of these traits with respect to PCA1.Likewise days to maturity, height and 1000 grain weight are on at the opposite end of right that indicates the role of these traits in this component.
Positioning/mapping of genotypes in the graph, indicates the tendency of relative grouping into three distinct groups.The ordination was mainly determined by yield related traits on axis one and, to a lesser extent by maturity elated traits on axis two.It is expected that selection of genotypes on the first component can lead to identification of high yield genotypes.Also, selection of genotypes in the An effort has been made to identify and introduce the traits that have high genetic heritability and have the greatest effect on the grain yield using the significant correlations as guiding principles between the measured traits and grain yield.It also helps the breeders to conduct indirect selection for important traits using relatively low-importance and easily measurable traits 14 .In the breeding programmes, selection is based on a large number of agricultural traits with an assumption that there may be positive and negative correlations between them.Therefore, statistical analyses that reduce the number of effective traits on yield are valuable for breeders.In this regard, the use of correlations between the traits is common, but these correlations The existence of such correlations allows the researcher to select genotypes indirectly and more accurately.Biological yield (q/ha) and days to heading have positive correlation with seed yield(q/ ha).Plant height and 1000 seed weight shows negative correlation with seed yield (q/ha).Similar results have been reported in earlier studies on wheat 17,18 Principal Component Analysis Bhanupriya et al.15 used 49 wheat and 2 triticale genotypes to study principal component analysis (PCA) in wheat indicated that five components (PC1 to PC5) accounted for about 75% of the total variation among traits in bread wheat cultivars.Out of total principal components retained PC1, PC2 and PC3 with values of 25.9%, 17.1% and 13.3% respectively contributed more to the total variation.Their study were rightly agree with the present investigation where PC1, PC2 and PC3 45.13%, 17.85 and 14.71% of the total variation, respectively.Similarly, Mohibullah et al.4 studied genetic diversity in wheat (Triticum aestivum L.) germplasm of one hundred found that the eigen values of three PCs out of ten, having 29.02%, 43.42% and 55.00% of the total variability with positive effects for most of the traits.Similar studies were performed in wheat by several researchers 5,[16][17][18][19][20][21][22] .In recent years, a large number of breeding lines have been created in different crops, and the selection of the best one in the regional preliminary experiments is very important.Multivariate statistical methods that can establish a relationship between the characteristics of cultivars can help grouping the cultivars and easily selecting them based on two-dimensional (biplots) and three-dimensional (triplots) diagrams.The principal component analysis, by summarizing the first-order correlated variables in the form of independent and finite components, enables the grouping of individuals in a two-dimensional or threedimensional space 23 .Some researchers have used the principal component analysis method to speed up the selection of genotypes in terms of the traits such as grain yield.Meanwhile, they have also taken advantage of this to reduce the cost of selection 24 .
In the two-dimensional and three-dimensional diagrams, which are based on the data derived from the principal component analysis, the effect of traits on the grouping of genotypes as different vectors and the location of each genotype is also shown based on the selected component type.The length of each vector represents its weight in creating distinct groups and is related to the component value for the corresponding trait.Genotypes can be compared by drawing a line perpendicular to the vector trait.In fact, the more the distance of the line of origin with the source is greater, the genotype function has a greater deviation than the average performance of the germplasm 25 .The angle between the vectors in these types of diagrams depicts their approximate correlation 26 .This type of relationship between agronomic traits and related vectors in the formed diagrams, categorizes the studied cultivars logically and directionally.This can be used as a factor in the selection of superior cultivars in the preliminary tests.Principal component analysis is useful for identifying diverse genotypes which can be used for future breeding programmes.In the present investigation first three PCA revealed the variability of 45.13, 17.85 and 14.71 percent respectively with the cumulative variance 77.69 of total phenotypic variance.Similar resuts were also documented in wheat and trticale. 15The PC1 is containing days to heading,biological yield and seed yield, which are contributing more in the variation.Our results are in agreement with earlier study reported in wheat. 16imilarly PC2 included days to maturity, plant height and 1000 grain weight as main components.These results were also reveled in earlier study on wheat. 21

Cluster Analysis
Nimbalkar et al. 27 conducted diversity analysis in 24 wheat cultivars for yield and yield contributing characters.The 24 cultivars were grouped into 12 clusters.The highest and lowest intracluster distances were observed in cluster III and I respectively.Dotlacil et al. 28 conducted the cluster analysis in 120 accessions of European winter wheat land races and obsolete cultivars.Eight groups of cultivars were identified, one of them was represented by the check cultivars.Very specific clusters 7 and 8 were composed of 14 cultivars from six countries.According to cluster analysis all the accessions were divided into three main groups A, B and C, which were further divided in to thirteen sub-groups4.The cluster IV and V were more clearly separated than cluster I, II and III.
The cluster analysis revealed that considerable variation existed among genotypes that could be implicated in selection of wheat for the development or improvement of cultivars and germplasm 18 .Similar clustering patterns were found out in the present study where twenty four wheat genotypes were evaluated for genetic diversity and four groups were formed.Supporting evidence from diverse studies in wheat augment our assertions in the present investigation, which further support the clustering pattern observed among these 24 gentoypes of wheat evaluated under temperate conditions 5,18,21,22 .In all their studies wheat genotypes were placed in different group according to their genetic similarity.Through these studies we can select diverse genotypes for breeding programme for producing transgressive segregants in minimum period of time.
The cluster analysis separated the 24 genotype into five groups based on the genetic variability.Similar results were reported in earlier studies on wheat including 24 and 120 genotypes 27,28

Conclusion
Genetic variability and diversity among the genotypes is a prime requisite for a successful and functional breeding programme.It is this variability, which is explored and exploited through different breeding strategies to bring desirable crop improvement as defined by the niche/ region specific or preference of improvement in the target trait.The developmental programme is further augmented and aided by the identification of desirable parents possessing significant genetic variability for the yield and yield attributing traits.The present investigation revealed presence of significant genetic variability and marginalized candidate wheat genotypes with specific importance for breeding wheat under temperate conditions of the valley.The genotypes were clustered into four distinct groups that showed sufficient variability available in the studied genotypes.The first three PCAs explained 45.13, 17.85 and 14.71 of total phenotypic variation, respectively.Promising genotyping with high yield per hectare and biological yield were identified.