Nội dung

Int.J.Curr.Microbiol.App.Sci (2021) 10(03): 129-140 International Journal of Current Microbiology and Applied Sciences ISSN: 2319-7706 Volume 10 Number 03 (2021) Journal homepage: http://www.ijcmas.com Original Research Article https://doi.org/10.20546/ijcmas.2021.1003.019 Principal Component Analysis of Soybean Genotypes under Post Anthesis Drought Stress Swati Saraswat* and Stuti Sharma Department of Plant Breeding & Genetics, Jawaharlal Nehru Krishi Vishwa Vidyalaya, Jabalpur, (M.P.), India *Corresponding author ABSTRACT Keywords PCA, Eigen value, Soybean, Principal component Article Info Accepted: 04 February 2021 Available Online: 10 March 2021 In the present study, PCA performed for phenological and yield component traits revealed that out of sixteen, only four principal components (PCs) exhibited more than 1.00 eigen value, and showed about 74.79 % total variability among the traits studied under stress condition while under normal condition out of sixteen, only five principal components (PCs) exhibited more than 1.00 eigen value and showed about 77.84% variability among the traits studied. Scree plot explained the percentage of variance associated with each principal component obtained by drawing a graph between eigen values and principal component numbers. A high value of PC score of a particular genotype in a particular PC denotes high value for those variables. On the basis of PCA analysis under both stress as well as normal condition soybean genotypes namely; TGX 852-3D, SQL 89 and YOUNG have been selected for all the yield component traits under study and CAT 2082 and MACS 58 have been selected for phenological traits. This will help in the further improvement of genotypes for post anthesis drought stress breeding. drought spells of various duration at different stages of crop particularly at seed fill stage is the main reason for soybean’s low productivity in India (ICAR-IISR Annual Report 2018-19). Introduction Soybean (Glycine max L. Merill) also known as golden bean is a legume species native to East Asia. Presence of symbiotic bacteria let them fix atmospheric nitrogen. Soybean is economically the most important bean in the world, providing vegetable protein for millions of people and ingredients for hundreds of chemical products. The enhancement of lower productivity of the crop is one of the major challenges in front of the soybean researchers. The climatic variability leading to delay in monsoon, For the development and growth of plants adequate water is needed. Oxidative stress and a reduction in photosynthetic characteristics are the consequences of less than optimal water (Guo et al., 2018). Drought affects soybean yield by affecting all stages of plant growth and development; from germination to flowering, and seed filling to 129 Int.J.Curr.Microbiol.App.Sci (2021) 10(03): 129-140 development as well as seed quality (Siddique et al., 2001; Manavalan et al., 2009). number of primary branches per plant, number of secondary branches per plant, number of pods per plant, number of pod clusters per plant, number of seeds per pod, number of seeds per plant, biological yield per plant, harvest index and seed yield per plant on a sample of three random plants per replication whereas for days to flowering and days to maturity, data were taken on whole plot basis. PCA was calculated using Ingebriston and Lyon (1985) method. Occurrence of drought stress during vegetative stage can be compensated with rains during later part of crop growth, however drought at terminal growth stage especially during seed filling to seed maturity stage would cause severe yield loss which could not be recovered by any means (Sionit and Kramer, 1977; Hirasawa et al., 1994; Saitoh et al., 1999). Terminal drought stress in soybean causes gradual reduction in photosynthetic rate, followed by senescence of leaves and reduced seed size that finally results in reduced grain yields (Brevedan and Egli, 2003; Manavalan et al., 2009). As several traits determine yield, we require a technique to identify and prioritize the important traits for effective selection. Principal component analysis, basically a well known data reduction technique identifies the minimum number of components, which can explain maximum variability out of the total variability (Anderson, 1972; Morrison, 1982) and also to rank genotypes on the basis of PC scores. Results and Discussion Under stress condition PCA performed for phenological and yield component traits in thirty soybean genotypes revealed that under stress condition, out of sixteen, only four principal components (PCs) exhibited more than 1 eigen value, and showed about 74.79 % total variability among the traits studied. So these four PCs were given due importance for further explanation (Table 1). Scree plot laid out between Eigen value and principal component showed total variation between them (Fig. 1). First principal component recorded the highest variation 30.49% (PC1) followed by 22.15% (PC2), 12.14% (PC3) and 9.73% (PC4). Total variation of four PCs was recorded to be 77.48%. Semi curve line obtained after fourth PC with little variation observed in each PC indicated that maximum variation was found in PC1, therefore, selection of lines for traits under PC1 may be desirable. Materials and Methods The present pot study was carried out at Glass House, Department of Plant Physiology, JNKVV, Jabalpur Madhya Pradesh during kharif 2018. The pot experiment was laid out in Completely Randomized Design with three replications. Thirty diverse genotypes of soybean were sown in pots inside glasshouse to screen them for drought tolerance and the genotypes were procured from ICAR-IISR (Indian Institute of Soybean Research), Indore and JNKVV released varieties from Department of Plant Breeding and Genetics, JNKVV, Jabalpur. Traits observed were days to flower initiation, days to full flowering, days to maturity, plant height at 30 days, plant height at maturity, number of nodes per plant, Rotated component matrix (Fig. 2) revealed that the first principal component (PC1) which accounted for the highest variation (30.49%) was mostly related with traits such as number of primary branches per plant, number of pods per plant, number of pod clusters per plant, number of seeds per plant, 130 Int.J.Curr.Microbiol.App.Sci (2021) 10(03): 129-140 biological yield per plant, 100 SW, harvest index, seed yield per plant. In second principal component (PC2) the traits days to flower initiation, days to 50 % flowering, days to maturity, number of nodes per plant, number of secondary branches per plant, while PC3 consisted of mainly two traits viz., plant height at 30 days and plant height at maturity whereas fourth principal component was related with number of seeds per pod (Table 2 and 3). On the basis of PCA, most of the important yield and yield attributing traits were present in PC1 and PC4. Rotated component matrix revealed that first four PCs are representing maximum variability (77.48%) hence, the traits falling to these four PCs may be given due importance in soybean drought breeding. AMS 26 A, CAT 2082, MACS 58, JS 21-73 and YOUNG (Table 4). Based on top PC scores genotypes were categorized in the table 5. Similar results were obtained by Iqbal (2008) for number of pods per plant, grain yield, biological yield per plant, 100 seed weight, harvest index, days to maturity and number of branches per plant and Ojo et al., (2012) for number of pods per plant, pod length, pod yield per plant, 100 seed weight and seed yield per plant. Under normal condition, out of sixteen, only five principal components (PCs) exhibited more than 1.00 eigen value and showed about 77.84% variability among the traits studied. Hence, these five principal components were given due importance for further explaination. PC1 includes SQL 89 which had highest PC score followed by TGX 852-3D, YOUNG, JS 21-72, SQL 8, HARDEE and SQL 31 indicated that these genotypes possesses high values of traits viz., number of primary branches per plant, number of pods per plant, number of pod clusters per plant, number of seeds per plant, biological yield per plant, 100 SW, harvest index, seed yield per plant which are mainly yield attributing traits. Scree plot had laid out between eigen value and principal component showed total variation between them (Fig. 3). First principal component recorded the highest variation 27.17% (PC1) followed by 19.59% (PC2), 13.24% (PC3), 9.69 % (PC4) and 7.774% (PC5). Total variation of five PCs was recorded to be 77.48%. Semi curve line obtained after fifth PC with little variation observed in each PC indicated that maximum variation was found in PC1; therefore, selection of lines for traits under PC1 may be desirable (Table 6). PC2 includes HARDEE which had highest PC score followed by CAT 3293, CAT 142, AGS 38, CAT 2082, MACS 58 AND YOUNG and it was mainly related with days to flower initiation, days to 50 % flowering, days to maturity, number of nodes per plant, number of secondary branches per plant. Rotated component matrix revealed that the first principal component (PC1) which accounted for the highest variation (27.17%) was mostly related with traits such as number of pods per plant, number of pod clusters per plant, number of seeds per pod, number of seeds per plant, biological yield per plant, harvest index, seed yield per plant. In second principal component (PC2) the traits viz., days to flowering, days to 50% flowering, days to The highest PC score was obtained by TGX 852-3D followed by SQL 89, SKY/AK-403, AMS 59, CAT 703, DAVIS and MACS 58 in PC3 for traits namely plant height at 30 days, plant height at maturity. PC4 recorded the highest value for traits viz., number of seeds per pod by the genotypes 131 Int.J.Curr.Microbiol.App.Sci (2021) 10(03): 129-140 maturity, number of secondary branches per plant. While PC3 consisted of mainly two traits viz., plant height at 30 days and plant height at maturity. Fourth principal component (PC4) was related with 100 SW. PC5 consisted of number of nodes per plant and number of secondary branches per plant (Table 7 and 8). genotypes possesses high values of traits viz., number of pods per plant, number of pod clusters per plant, number of seeds per pod, number of seeds per plant, biological yield per plant, harvest index, seed yield per plant which are mainly yield attributing traits. The highest PC score of CAT 142 followed by CAT 3293, AGS 38, MACS 58, CAT 649, CAT 2082, CAT 703, HARDEE, SQL 88 and YOUNG in PC2 was mainly related with days to flowering, days to 50% flowering, days to maturity, number of secondary branches per plant. On the basis of PCA, most of the important yield and yield attributing traits were present in PC1 and PC5. Rotated component matrix revealed that first five PCs are representing maximum variability (77.48%) hence, the traits falling to these five PCs may be given due importance in soybean breeding. The highest PC score was obtained by AMS 59 followed by YOUNG, JS 21-17, AMS 19 B, TGX 852-3D, JS 20-29 and MACS 58 in PC3 for traits namely plant height at 30 days and plant height at maturity (Fig. 4). TGX 852-3D had the highest PC score followed by CAT 2082, CAT 3293, DAVIS, SQL 89, JS 20-29 in PC1 indicated that these Table.1 Eigen values, percentage of total variation and cumulative percentage for corresponding sixteen traits under Stress condition Traits Eigen value DFI DFF DM Pl Ht. at 30 DAS Pl Ht. at maturity NNP NPBP NSBP NPP NPCP NS/ pod NSP BY 100 SW HI Principal component (PC) PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 PC14 PC15 SYPP PC16 132 4.879 3.545 1.986 1.557 0.780 0.726 0.651 0.543 0.514 0.290 0.249 0.124 0.095 0.038 0.019 Variability (%) 30.495 22.158 12.414 9.730 4.873 4.535 4.069 3.392 3.213 1.814 1.558 0.776 0.592 0.240 0.119 Cumulative % 30.495 52.653 65.067 74.797 79.670 84.204 88.273 91.666 94.879 96.693 98.251 99.026 99.618 99.858 99.977 0.004 0.023 100.00 Int.J.Curr.Microbiol.App.Sci (2021) 10(03): 129-140 Table.2 Principal Components for 16 phenological and yield contributing traits of soybean genotypes under stress condition Traits DFI DFF DM Pl Ht. at 30 DAS Pl Ht. at maturity NNP NPBP NSBP NPP NPCP NS/ pod NSP BY 100 SW HI SYPP Principal Components PC2 PC3 -0.207 0.737 -0.282 0.762 -0.303 0.662 -0.102 0.916 0.505 0.691 0.132 0.667 0.411 0.001 0.065 0.582 0.242 0.222 0.167 0.193 -0.552 0.003 -0.285 0.322 0.086 -0.180 -0.436 -0.453 -0.270 0.000 -0.150 -0.074 PC1 0.050 0.127 -0.155 -0.080 -0.121 0.112 0.470 0.203 0.874 0.878 -0.030 0.790 0.844 0.599 0.699 0.906 PC4 -0.207 -0.282 -0.303 0.916 0.671 0.132 0.001 0.065 0.222 0.193 0.003 0.322 -0.180 -0.453 0.000 -0.074 TRAITS Table.3 Interpretation of rotated component matrix for the traits having values >1 in each PCs under stress condition PC 1 No of primary branches per plant No of pods per plant No of pod clusters per plant No of seeds per plant Biological yield per plant 100 seed weight Harvest Index Seed yield per plant PC 2 Day to flowering PC 3 Plant height at 30 days PC 4 No of seeds per pod Days to 50 % flowering Days to maturity Plant height at 60 days - - No of nodes per plant No of secondary branches per plant - - - - - - - - - - 133 - Int.J.Curr.Microbiol.App.Sci (2021) 10(03): 129-140 Table.4 PCA scores of soybean genotypes under Stress condition S.No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Genotypes JS 20-29 JS 20-69 JS 20-98 JS 97-52 DAVIS YOUNG JS 21-17 AMS MB 518 TGX 852-3D MACS 58 SKY/AK-403 HARDEE JS 21-73 CAT 142 CAT 649 CAT 703 CAT 3293 CAT 2082 AGS 38 AMS 59 AMS 19 B AMS 26 A AMS 148 SQL 8 SQL 31 SQL 88 SQL 89 SQL 106 JS 21-71 JS21-72 PC1 0.206 -0.168 -1.406 0.427 0.401 3.228 0.837 -0.253 3.403 -0.524 -2.231 2.280 -0.776 -1.228 -1.708 -1.970 0.839 0.965 -2.204 -4.168 -3.200 -1.692 0.177 2.605 2.197 -0.840 6.097 -2.487 -1.462 2.654 PC2 0.117 -0.400 -0.416 0.382 0.174 1.168 -1.685 -1.134 -1.552 1.712 -0.901 4.348 -1.483 3.429 -1.496 0.589 4.112 2.226 2.617 -0.152 0.097 -2.904 -0.417 -3.070 0.996 -0.173 -0.934 -1.374 -1.360 -2.516 PC3 0.937 -1.801 -0.344 -0.157 1.288 -1.580 0.489 -0.253 3.516 1.120 2.106 0.436 -0.441 0.294 0.601 1.405 -0.567 -1.376 -1.349 1.516 0.813 -0.185 0.117 -2.128 -1.380 -1.214 2.670 -1.085 -1.090 -2.359 PC4 0.362 -0.429 0.746 0.383 0.368 1.197 0.092 0.501 0.831 1.729 -0.646 -0.303 1.373 0.561 0.877 0.184 0.009 2.145 -0.573 -1.845 -2.315 2.602 0.478 -0.705 -3.156 -0.520 -1.805 -0.715 -1.197 -0.225 Table.5 List of selected genotypes in each principal component under Stress condition PC1 SQL 89 TGX 852-3D YOUNG JS21-72 SQL 8 HARDEE SQL 31 PC2 HARDEE CAT 3293 CAT 142 AGS 38 CAT 2082 MACS 58 YOUNG PC3 TGX 852-3D SQL 89 SKY/AK-403 AMS 59 CAT 703 DAVIS MACS 58 134 PC4 AMS 26 A CAT 2082 MACS 58 JS 21-73 YOUNG Int.J.Curr.Microbiol.App.Sci (2021) 10(03): 129-140 Table.6 Eigen values, percentage of total variation and cumulative percentage for corresponding sixteen traits in soybean genotypes under Normal condition Traits DFI DFF DM Pl Ht. at 30 DAS Pl Ht. at maturity NNP NPBP NSBP NPP NPCP NS/ pod NSP BY 100 SW HI SYPP Principal component (PC) PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12 PC13 PC14 PC15 PC16 Eigen value 4.348 3.135 2.120 1.551 1.244 0.841 0.748 0.561 0.451 0.413 0.241 0.173 0.068 0.064 0.037 0.006 Variability (%) 27.177 19.593 13.248 9.691 7.774 5.257 4.678 3.508 2.817 2.580 1.505 1.084 0.424 0.397 0.230 0.035 Cumulative % 27.177 46.770 60.019 69.710 77.845 82.742 87.420 90.928 93.744 96.325 97.829 98.914 99.338 99.735 99.965 100.00 Table.7 Principal Components for 16 phenological and yield contributing traits of soybean genotypes under normal condition Traits DFI DFF DM Pl Ht. at 30 DAS Pl Ht. at maturity NNP NPBP NSBP NPP NPCP NS/ pod NSP BY 100 SW HI SYPP Principal Components PC2 PC3 PC4 -0.009 0.449 0.690 -0.113 0.496 0.653 -0.345 -0.069 0.675 -0.042 -0.089 0.917 0.389 0.158 0.815 0.451 -0.004 -0.400 0.352 -0.261 0.223 0.089 -0.272 0.634 0.119 -0.190 -0.277 0.253 -0.212 -0.410 -0.426 0.255 0.374 -0.231 0.045 -0.112 0.224 0.083 0.427 -0.454 -0.494 0.417 -0.486 -0.078 -0.063 -0.296 0.046 0.153 PC1 0.322 0.394 0.039 0.071 0.078 0.394 -0.225 -0.136 0.864 0.746 0.436 0.882 0.571 -0.055 0.710 0.865 135 PC5 -0.060 -0.221 0.006 0.069 0.034 0.500 0.748 0.258 -0.150 -0.171 0.366 0.028 -0.163 0.149 0.242 0.124 Int.J.Curr.Microbiol.App.Sci (2021) 10(03): 129-140 TRAITS Table8 Interpretation of rotated component matrix for the traits having values >01 in each PCs under Normal condition PC 1 Number of pods per plant PC 2 Days to flowering Number of pod clusters per plant Number of seeds per pod Number of seeds per plant Days to 50% flowering Biological yield per plant Harvest Index Seed yield per plant Days to maturity Number of secondary branches per plant - PC 3 Plant height at 30 days Plant height at 60 days - PC 4 100 seed weight - PC5 Number of nodes per plant Number of secondary branches per plant - - - - Table.9 PC scores of soybean genotypes under Control condition S.No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Genotypes JS 20-29 JS 20-69 JS 20-98 JS 97-52 DAVIS YOUNG JS 21-17 AMS MB 518 TGX 852-3D MACS 58 SKY/AK-403 HARDEE JS 21-73 CAT 142 CAT 649 CAT 703 CAT 3293 CAT 2082 AGS 38 AMS 59 AMS 19 B AMS 26 A AMS 148 SQL 8 SQL 31 SQL 88 SQL 89 SQL 106 JS 21-71 JS21-72 PC1 1.424 -0.331 0.182 -0.516 3.182 -0.674 0.986 -0.472 6.185 -0.485 -0.801 0.239 0.769 0.903 0.486 -0.787 3.232 3.576 -2.032 -3.804 -3.108 -0.388 0.153 -2.515 -1.475 -0.075 1.892 -1.858 -2.065 -1.824 PC2 -0.588 0.228 0.393 -0.407 -0.465 1.039 -1.924 -0.836 -3.185 1.947 -1.114 1.123 -2.648 3.431 1.777 1.323 3.331 1.573 3.045 0.223 0.698 -1.255 -0.506 -2.291 -0.043 1.074 -0.501 0.112 -2.470 -3.086 136 PC3 1.443 -0.048 0.527 0.236 -0.561 2.640 2.338 0.200 1.714 1.288 0.919 -0.529 -0.150 0.580 0.240 0.277 -1.428 -1.978 -1.244 2.867 1.892 -0.130 0.756 -2.585 -2.406 -0.840 -1.559 -1.607 -1.321 -1.532 PC4 0.359 1.073 1.213 0.358 -3.481 2.308 1.887 -0.554 -1.521 0.598 -1.448 1.554 0.249 -0.236 -1.106 -1.231 0.294 0.810 -0.638 -2.118 -0.935 0.876 0.436 -0.369 -1.191 0.461 1.121 -0.369 1.236 0.364 PC5 0.131 1.735 2.447 0.206 1.019 -1.150 0.433 0.464 -0.020 1.460 -1.086 -0.276 -1.160 -0.788 -0.489 -0.269 -2.428 -0.388 0.502 -0.117 -1.385 1.455 0.325 0.846 0.829 1.327 -0.039 -0.115 -2.048 -1.418 Int.J.Curr.Microbiol.App.Sci (2021) 10(03): 129-140 Table.10 List of selected genotypes in each principal component under Normal condition S.No. 1 2 3 4 5 6 7 8 9 10 PC1 TGX 852 -3D CAT 2082 CAT 3293 DAVIS SQL 89 JS 20-29 PC2 CAT 142 CAT 3293 AGS 38 MACS 58 CAT 649 CAT 2082 CAT 703 HARDEE SQL 88 YOUNG PC3 AMS 59 YOUNG JS 21-17 AMS 19 B TGX 852-3D JS 20-29 MACS 58 PC4 YOUNG JS 21-17 HARDEE JS 21-71 JS 20-98 SQL 89 JS 20-69 PC5 JS 20-98 JS 20-69 MACS 58 AMS 26 A SQL 88 DAVIS Fig.1 Scree plot of principal component analysis of soybean genotype between eigen value and principal components under stress condition Fig.2 Phenlogical and yield traits of soybean genotypes under stress condition shown in Bar Diagram 137 Int.J.Curr.Microbiol.App.Sci (2021) 10(03): 129-140 Fig.3 Scree plot of principal component analysis of soybean genotype between eigen value and principal components under normal condition Fig.4 Phenlogical and yield traits of soybean genotypes under normal condition shown in Bar Diagram PC scores in PC4 were recorded the highest value for traits viz., 100 SW. by the genotypes YOUNG, JS 1-17, HARDEE, JS 21-71, JS 20-98, SQ 89 and JS 20-69. al., 2018. for days to 50% flowering, days to maturity, plant height, number of branches per plant, number of nodes per plant, number of pods per plant, number of pods per node, number of seeds per plant, biological yield per plant and seed yield per plant. However, JS 20-98, JS 20-69, MACS 58, AMS 26 A and SQL 88 had the highest PC scores in PC5 for number of nodes per plant and number of secondary branches per plant. From the above discussion under stress condition it is clear that SQL 89 holds the first position followed by HARDEE and TGX 852-3D on the basis of PC score in all principal components. When we considered the entire PC with PC scores and character basis then SQL 89 ranked first because it is present in PC1 as well as in PC3. SQL 89 contributes maximum character because most Based on top PC scores genotypes were categorized in the Table 10. Similar results have been found by Badkul et al., 2014 for plant height, number of branches per plant and yield per plant and by Dubey et 138