Current IssueSpecial IssueEditorial BoardOnline First ArticlesArchive
Current IssueSpecial IssueEditorial BoardOnline First ArticlesArchive

Full Research Article

Stability Analysis of Greengram [Vigna radiata (L.) Wilczek] Genotypes for Seed Yield in Telangana using Regression, AMMI, GGE and WAAS Biplot Models

K
K. Rukminidevi1,*
0009-0004-6638-0920
A
A. Saritha1
K
K. Parimala1
G
G. Eswara Reddy1
A
A. Dinesh1
N
N. Lingaiah1
B
B. Supriya1
1Agricultural Research Station, Professor Jayashankar Telangana Agricultural University, Madhira-507 203, Telangana, India.

ABSTRACT

Background: To increase area and production of greengram, breeders must develop high yielding, stable and adaptable varieties. This study aims to identify such superior greengram genotypes that perform consistently well across different locations.

Methods: The experiment was carried out during Kharif, 2023 in six locations at different Agricultural Research Stations of Telangana state. A total of 15 genotypes were evaluated in RBD with three replications. The data on seed yield was subjected to statistical analysis. The G×E interaction was studied as per Eberhart and Russel model, AMMI, GGE biplot and WAAS model analysis.

Result: Regression analysis revealed that genotypes MGG-385 and MGG-573 considered as stable. AMMI analysis of variance indicated that genotypes contributed 13.1% followed by environment (55.4%) and G×E interaction (20.48%). With further examination of GGE using AMMI analysis, two significant principal components were separated explaining 69.7% of variance interaction (PCI-42% and PC2-27.7%). The results of AMMI 1 and 2 biplot analysis, the genotypes MGG-385 and MGG-295 were considered as stable with high mean seed yield and recorded nearly zero IPCA1 score. The environments of Madhira and Tandur had recorded the lowest IPCA1 scores had less interaction effects. AMMI stability value and stability indexes were indicated MGG-385 was more stable. Y×WAAS graph revealed the genotypes VBN-4, MGG-570, MGG-571 were unstable had high seed yield, while the environments Palem, Tornala and Warangal provided high seed yield and presented good discrimination ability as had recorded high WAAS values. The genotypes MGG-573, MGG-556, MGG-385 and MGG-564 were broadly adopted. Based on all the stability models, the genotype MGG-385 was considered as stable with high mean seed yield across locations.

INTRODUCTION

Pulses are vital to india’s food and nutritional security offering a sustainable path for soil improvement and environmental health. Greengram [Vigna radiata (L.) Wilczek] plays a transformative role in diversifying Indian agriculture and addressing malnutrition in a largely vegetarian population. It holds significance as the third most important pulse crop in India. It provides an affordable source of nutrients, containing 24-26% protein, carbohydrates and essential minerals and vitamins. Furthermore, it plays a vital role in soil health through biological nitrogen fixation. As a centre pillar of the nations “pulses revolution”, mungbean has outperformed all other pulse crops, achieving a staggering growth rate of over 185% during the last one and half decade. Globally, Asia produces 90% of the mungbean crop, with India as the top producer, cultivating it over 3.7 million hectares with a production of 4.23 million tons. The development of short duration, photo thermo insensitive and disease resistant varieties has expanded their reach allowing them to thrive as Spring/Summer crops in north India and as vital relay crops in the rice fallows of the southern peninsula. (AICRP Kharif Pulses-Annual Report, 2025-26). This low and unstable production is primarily due to the inconsistent performance of varieties under varied environmental conditions. Therefore, a critical breeding goal, especially in the context of climate change adaptation, is to develop varieties that are less sensitive to variations in weather conditions to stabilize production (Sidramappa et al., 2024). An effective cultivar must possess a combination of both high yield and genetic stability (Yihunie and Gesesse, 2018). The challenge lies in the Genotype × Environment (G×E) interaction, which is often non-additive, meaning a genotype’s performance largely depends on the specific environment. Identifying cultivars with predictable performance requires a rigorous analysis of adaptability and stability (Pillai et al., 2010).  Stability analysis methods are necessary to recommend genotypes that consistently outperform and yield higher across various locations, seasons  and environments. Selecting robust and suitable methods for studying stability is a critical and challenging task for any breeder in a varietal development program (Piepho, 1996).
       
Several models have been developed to dissect and address the G×E interaction. The Eberhart and Russell (1966) model suggested considering both linear (bi) and non-linear (S2di) components of the GxE interaction to determine a genotype’s phenotypic stability. The AMMI model is a hybrid analysis that uniquely separates main (additive) and interaction (multiplicative) effects (Gauch, 1992).  It uses principal component analysis (PCA) on the interaction portion to explain the G×E pattern in greater detail. AMMI also provides a visual representation via a biplot using the first Interaction Principal Component Axis (IPCA1) and mean yield to diagnose interaction patterns and identify stable genotypes (Gauch, 2006). The GGE biplot analysis is widely employed to detect and graphically indicate the GEI across multiple environmental trials (Yan et al., 2000). It is based on the first and second symmetrically scaled principal components derived from the Singular Value Decomposition of environment-centered data. GGE biplots are effective tools for facilitating variety evaluation and the delineation of mega-environments. WAAS is a novel technique that offers a superior framework by combining and improving upon AMMI and GGE (Olivoto and Lucio, 2020). Unlike AMMI, which often relies only on the first PC, WAAS incorporates all Principal Components (PCs) from the AMMI ANOVA, thus effectively capturing the entire variance of GEI. The WAAS biplot is used for the joint interpretation of a genotype’s average yield performance and stability across environments. To develop an effective selection strategy for future greengram breeding programs, this study used multivariate statistical techniques to identify stable and superior greengram genotypes, aiming to upscale cultivation and ensure consistent output.

MATERIALS AND METHODS

The experiment was conducted with 15 elite greengram genotypes during Kharif season, 2023 at six locations comprising of different Agroclimatic zones working under PJTAU, Telangana state, India (Table 1). The experimental material was planted in a RBD with three replications in each location adopting row to row and plant to plant spacing of 30 cm and 10 cm respectively. All the recommended package of practices and plant protection measures were followed to raise a good crop. The data on seed yield (kg/ha) was subjected to statistical analysis. The G×E interaction was studied i.e., Eberhart and Russel, (1966) analysis were carried out using the software P.B. tools (ver.1.3) developed by International Rice Research Institute Philippines. Additive main effects and multiplicative interaction (AMMI) (Gauch and Zobel, 1996), GGE biplot or site Regression model (Yan and Kang, 2003) and WAAS (Olivoto and Lucio, 2020) analyses were carried out using the software R package for multi environment trial analysis.

Table 1: Details of greengram genotypes along with environments.

RESULTS AND DISCUSSION

Analysed data of each environment indicated that genotypes were found to be significant in all the six environments revealed greater magnitude of genetic variation for seed yield. Pooled analysis of variance for seed yield (Table 2) indicated that the genotypes, environments and genotype x environment were significant indicating the presence of variation among genotypes, environments and influence of G×E interaction.

Table 2: Pooled analysis of variance of seed yield in greengram genotypes.


       
Both linear and non linear components were found to be significant indicating the presence of both stable and non stable genotypes which was also confirmed by Manivannan et al. (2023). Stability parameters such as mean (xi), regression coefficient (bi) and deviation from regression (S2di) are considered to assess the stability of different genotypes for seed yield as suggested by Eberhart and Russel, (1966). The genotypes were categorized into three groups high yielders (939-1136 kg/ha), medium yield (827-880 kg/ha) and low yielders (742-811 kg/ha). Among the genotypes, MGG-556, MGG-570, MGG-564 and MGG-571 recorded high mean yield, regression coefficient more than unity (bi>1) and non significant S2di value indicating that these varieties perform better under good environment and were stable (Table 3). The genotype G9 (MGG-385), G4 (MGG-573) has recorded high yield, regression coefficient (bi=1) and non-significant S2di values, indicating these genotypes have an average response to environment and stable due to predictable nature and perform better across environments. The entries G13(MGG-295) and G6(MGG-576) and G1(MGG-563) recorded medium yield, bi=1 and non-significant S2di can perform better under poor environments and stable. The genotype G15 (VBN-4) has recorded bi value nearer to unity and significant S2di value indicating that this variety is suitable for good environment and unpredictable. The genotypes IPM-2-14, WGG-42, MGG-562 and MGG-565 were poor yielders and recorded non significant S2d and less regression coefficient (bi<1), hence seed yield did not improve with the improvement in the environment. However, due to the fact that the G×E conserts of both additive and multiplicative model, the recent researchers used the AMMI, GEE biplot and WAAS models. Hence the present data was used to analyze both models to compare the results.

Table 3: Genotypic means with stability parametres for seed yield (kg/ha) in greengram genotypes.


       
The analysis of variance for AMMI analysis elucidated that the significant difference existed among genotypes, environments and genotypes x environment interaction and witnessed the considerable influence of environments and interaction of genotypes with environments in expressing of the grain yield (Table 4). Further, genotypes contributed 13.18%, environments contribute (55.47%) more in total sum of squares followed by genotype x environment interaction (20.48%) revealing that genotypic architecture of the genotype, environment and their interaction in manifestation of the trait. The larger portion of the yield variation explained by environments indicated that the environments causing the most of the variation for seed yield. The seed yield over environments ranged from 588 kg/ha in Karimnagar (E1) to 1294 kg/ha in Tornala (E4). A large proportion of yield variation explained by environments indicated thet the environments were diverse with large difference among environments causing the most of the variation for grain yield. The genotypic seed yield ranged from 743 kg/ha G11(IPM 2-14) to 1136 kg/ha G7(MGG-556). G×E interaction was a crossover type with different yield ranking of genotypes across environments. The significant interaction in genotype and environment for yield validated the need to take more care while selecting the promising genotypes by considering stability and adaptability. Significant differences across locations were also earlier reported by Sanasam et al. (2025) using AMMI model. With further division of GGE using AMMI analysis, two significant principal components were separated explaining 69.7% of variance interaction (PCI 42% and PC2 27.7%). Earlier reports confirmed that in most of the cases the maximum genotype and environment interaction could be explained through using the first three PCA’s. Naresh et al. (2025); Sanasam et al. (2025) in greengram, Chandramohan et al. (2023) in rice evaluated the interactions between genotypes and environmental axes IPCA1 and IPCA2. While Mahalingam et al. (2019) reported IPCA1 alone may decide the G×E interaction in this study, whereas Manivannan et al. (2023) also reported that IPCA1, IPCA2 and IPAC3 contribute 46.3%, 37.5% and 16.0% towards interactions respectively.

Table 4: AMMI and GGE biplot analysis of variance for seed yield in greengram genotypes.


       
The results of the AMMI analysis further enlightened the relative contribution of the first two IPCA axes to the interactive effects by plotting with genotype and environment means as presented in Fig 1. In the biplot, environments are designated by the letter E followed by number 1 to 6 suffix (E1-E6) while, genotypes represented by numbers for 1 to 15 (G1-G15). When a variety and environment have the same sign on PCA1 axes their interaction is positive and if opposite their interaction is negative. Thus if a variety has a IPCAI score near to zero, it has small interaction effect and committed to the stable over wide environments. Conversely, varieties with high mean yield and large IPCA1 scores were considered as explicitly adapted to specific environments. Similar views were expressed in their studies by Manivannan et al., (2023); Shobanadevi et al., (2023); Naresh et al. (2025) and Sanasam et al. (2025).

Fig 1: AMMI-I biplot mean seed yield (kg/ha) of the genotypes (G) and environment (E) against their respective IPCA1 (Y-axis) scores.


       
The greengram genotypes G7(MGG-556) recorded high seed yield followed by G12 (MGG-570), G15 (VBN-4), G9(MGG 385) and G8 (MGG-571). In the present study G7(MGG556) has recorded highest seed yield value (horizontal axis) with positive IPCA1 score (vertical axis) of 9.91 followed by G12 (MGG-570) and G14 (MGG-564) indicating positive genotype × environmental interaction, hence these genotypes can be recommended for cultivation under favorable environments. The genotype G9(MGG-385) have recorded nearly zero IPCA1 score (0.03) (Table 5) indicating minimal interaction with the environment in which genotype showed above average yield and identified as most stable genotype with less yield variation across different environments. The other genotypes had less than the mean seed yield and found specific adoption to few tested environments. Most of the genotypic stability was attributed by environments E2 (Madhira) (-2.74) and E4 (Tandur)(-5.92) which had the lowest IPCA 1 scores had very little interaction effects and thus the performance of all genotypes in these environments was quite well. Hence these were considered as favorable environments for all the tested genotypes. With respect to GEI contribution E5 (Tornala) (17.84) and E3 (Palem) (13.99) exhibited the highest level of contribution as indicated by their high IPCA1 scores had the greatest effect on the G×E, so the comparative genotypes ranking in these environments were unstable.

Table 5: Ranking of Genotypes according to yield and stability indexes.


       
The environment having the long spokes is said to be more interactive (Fig 2) and among the environments E2 (Madhira) and E5 (Tornala) has the long spokes than E6 (Warangal), E4 (Tandur) and E3 (Palem) environments considering the genotypes. Those genotypes are nearer to origin are considered to be less interactive with environments. Based on this criterion genotypes G9 (MGG 385) within the orient and G13 (MGG295) close to the biplot origin had a smaller spoke of G´E were considered as less interactive genotypes with environments and generally stable in all environments. The genotypes G7(MGG-556), G12(MGG-570), G8(MGG-571) and G14(MGG-564) are far from the origin had the most variation due to environmental changes.

Fig 2: AMMI-2 biplot (PCA1 on X-axis and PCA2 on Y-axis).


       
Based on AMMI biplot 1 and 2, the genotype MGG-385 was identified as stable genotype with high seed yield. E4 (Tandur) and E3 (Palem) environments where the vectors were shortest and could discriminate between genotypes. In this graph, sites with short spokes did not exert strong interactive forces. Those with long spokes exerted strong interaction.
       
To get a clear cut ranking of the genotypes regarding stability, AMMI stability value (ASV) stability indexes were computed (Table 5). G9 (MGG-385) showed the minimum ASV (2.117) value and ranked as number one regarding stability, while G8 (MGG 571) has showed highest ASV (21.83) value and showed poorest stability.
       
GGE biplot helps in identifying GxE interaction pattern of data and provides us with clear picture of which genotype perform best in which environments and thus facilitates mega environment identification than AMMI (Table 4). Otherwise, both GGE and AMMI models are equivalent as far as their accuracy is considered (Gurmu et al., 2012). GGE biplot analysis also enables visual assessment of adaptability and stability. GGE biplot is presented with two principal components explaining a total of 75%. GGE variation (PC1=55.53%; PC2=19.9% represented in Fig 3. The first Principal component is represented on the X-axis and across its values is estimated yield i.e., genotypes that have higher PC1 values are considered to be more productive. The second principle component is represented on the Y-axis and presents the stability of the genotypes. Estimation of yield and stability of genotype was done by using so called AEC (average coordinates of the environment) method (Yan, 2001; Yan and Hunt, 2001). By projecting the genotypes on AEA axis, the genotypes are ranked by yield where the yield increases in the direction of arrow. Single arrowed lines in the Average Environment Coordination Abscissa (AEC) points to higher mean seed yield across environments (Fig 4). The double arrowed line in the AEC Coordinate representing highest variation in either direction. Hence, the genotypes G9 (MGG-385) and G13 (MGG-295) are stable, whereas others are highly interactive with environments. Hence, the genotypes G9 (MGG-385) and G13 (MGG-295) are found to be ideal with high mean and stability.

Fig 4: Biplot of stability and mean performance of genotypes across environments.


       
Discriminating and representativeness are the most important parameters of the GGE biplot when evaluating an environment. In Yan and Tinker, (2006) model a long environmental vector had high discriminating ability and short one had low discrimination therefore as shown in Fig 5 among tested locations E2(Madhira) and E5(Tornala) were identified as the potential environments for discriminating ability and representativeness. Yan et al. (2000); Laxuman et al., (2025) and Sanasam et al., (2025) also emphasized that the environments with long vectors are more discriminating and representativeness for consideration in future studies. Among the environmental vectors E3-E5, E2-E4, E2-E6 are found to be positively correlated (acute angle) E1-E6, E1-E2, E1-E3 vectors from obtuse angle and hence they are negatively correlated. E3-E6 form right angle and hence not correlated between them. Longer the distance between environmental vectors indicates that they are dissimilar in discriminating the genotypes. From the Fig 5, E2-E4 form one group whereas E1-E4 form two different groups. Environmental vector of E5 is the longest indicating that it is the environment with most discriminating ability followed by E2 (Madhira) and E3 (Palem) while E1 (Karimnagar) and E4 (Tandur) is least discriminating environments.

Fig 5: GGE biplot –Discriminativeness and representativeness of genotypes for seed yield in greengram.


       
What won where view of the GGE biplot is the best model for multi environment data for classifying the environments and also to select best performing genotypes in each (Yan et al., 2000). The equality line divides biplot into different sectors and winning genotypes are located on vertex of each sector. Genotypes located on the vertices of the polygon performs either the best or the poorest in one or the more environments. Vertex genotypes have longest vectors in their respective direction which is a measure of the responsiveness of the environment. The vertex genotypes G1 (MGG-563), G8 (MGG-571), G12 (MGG-570), G15 (VBN-4), G11 (IPM 2-14), G3 (MGG-565) and G5 (WGG-42) have longest vectors in their respective directions (Fig 6). The genotypes G9 (MGG-385) and G13 (MGG-295) considered as stable and performed better under all the environments. The genotypes G3 (MGG-565), G5 (WGG-42) and G11 (IPM 2-14) were best performers in mega environment consisting in E1(Karimnagar) while G1 (MGG-563), G8 (MGG-571) and G12 (MGG-570) perform well in mega environments consists of E5(Tornala), E3 (Palem) and E4 (Tandur) and G15(VBN-4) in E1 (Karimnagar) and E6 (Warangal) are the winning genotypes. The other genotypes G4 (MGG-573), G14 (MGG-564), G12 (MGG-570), G6(MGG-576) and G2(MGG-553) are poorest performers of all the environments because there is no environment in their sector. Hence for different mega environments, different genotypes have to be selected and deployed for each.

Fig 6: What-won-Where GGE Biplot for yield (kg/ha).


       
Y×WAAS biplot was divided into four quadrants which represented the classification of genotypes and environments (Fig 7). The quadrants in the graph represent (QI and QII) higher mean, (QIII and QIV) lower means. (Q1 and QIV) positive IPCA1 and (QII and QIII) negative IPCAI scores (Table 5). G3(MGG-565) and G1(MGG-563) grouped in the I quadrant were unstable genotypes with productivity below the grand mean and E1 environment was included with high discrimination ability as had high WAAS value. In quadrant II, G15 (VBN-4), G12 (MGG-570) and G8 (MGG-571) were present which were unstable genotypes, but had seed yield above the highest average value. The environments E2(Madhira), E3 (Palem), E5(Tornala) and E6(Warangal) included in the II quadrant provides high seed yield, they presented a good discrimination ability as had high WAAS values except E2 (Madhira). Genotypes G2 (MGG-553), G5 (WGG-42),G6 (MGG-576),G10 (MGG-562),G11 (IPM-2-14), G13 (MGG-295) included in the III quadrant had a low seed yield, low WAAS values so considered stable but no environment with high discrimination ability was identified. The genotypes G4 (MGG-573), G7 (MGG-556), G9 (MGG-385) and G14 (MGG-564) present in the IV quadrant were broadly adopted but their performance regarding yield were just nearer to mean. AMMI biplot found E2(Madhira) and E4 (Tandur) as favorable while E5 (Tornala) and E6 (Warangal) with the greatest GEI as unfavorable for testing. However, Y×WAAS grouped the environment only based on discriminating ability, so in this biplot E3 (Palem), E5 (Tornala) and E6 (Warangal) all grouped in II quadrant with good discriminating ability. Further, these environments have high WAAS values indicating that they have higher interactions with the studied genotypes. The higher mean performance of these environments regarding seed yield indicated that their influence on the yield was positive. Similar results regarding GEI from the WAAS biplot were also observed for chickpea genotypes by Arshad et al., (2024).

Fig 7: Y×WAAS biplot for seed yield (kg/ha) on X-axis and weighted average of absolute scores on Y-axis.

CONCLUSION

Yield stability is significantly influenced by both environment and genotypes, with their major interaction effects defined by the first two components of the AMMI model. Two genotypes were identified as stable with above-average yield: G9 (MGG-385) (960 kg/ha), G13 (MGG-295) (880 kg/ha) and had nearly zero IPCA1 scores. The genotypic stability was largely driven by environments E2 (Madhira) and E4 (Tandur), which were favorable for high seed yield expression across most of the genotypes. While E5 (Tornala) recorded the highest mean seed yield (1294 kg/ha) and was the most ideal environment, E2 (Madhira) and E5 (Tornala) were the most representative. G9(MGG-385) specifically exhibited the best combination of high seed yield and stability across all environments. Multivariate models are thus beneficial for identifying stable genotypes in multi-location testing.

ACKNOWLEDGEMENT

It is a part of regular research work and authors are duly acknowledge PJTAU for providing financial and research facilities.
 
Disclaimers
 
The views and conclusions expressed in this article are solely those of the authors and do not necessarily represent the views of their affiliated institutions. The authors are responsible for the accuracy and completeness of the information provided, but do not accept any liability for any direct or indirect losses resulting from the use of the content.
 
Informed consent
 
Informed consent is not applicable as work is a crop based plant research.

CONFLICT OF INTEREST

The authors declare that there are no conflicts of interest.

REFERENCES


  1. Annual Report-Vol-II (2025-26). ICAR-All India Coordinated Research Project on Kharif Pulses, IIPR, Khanpur.

  2. Arshad, W., Aleem, S., Khan, M.I., Saqib, M., Zeeshan, M., Mohsin, M.U., Malik, A., Sharif, I., Younas, A., Ihsan, F. and Tauseef, M. (2024). AMMI, BLUP and WAASB stability models for the selection of stable chickpea (Cicer arietinum L.) genotypes in different agroecological regions of Pakistan. Journal of Food and Agricultural Technology Research. 3(02): 66-81.

  3. Laxuman, Avinalappa, H., Kuchanur, P.H., Kumar, K.S. and Kumar, G.A. (2025). Seed yield stability assessment of chickpea genotypes through AMMI and Bi plot analysis. Legume Research-An International Journal. 48(2): 218-223. doi: 10.18805/LR-4911.

  4. Chandramohan, Y., Krishna, L., Srinivas, B., Rukmini, K., Sreedhar, S., Prasad, K.S., Kishore, N.S., Rani, C.V.D., Singh, T.V.J. and Jagadeeshwar. R. (2023). Stability analysis of short duration rice genotypes in Telangana using AMMI and GGE Bi-plot models. Environment Conservation Journal. 24(1): 243-252.

  5. Eberhart, S.A. and Russel, W.A. (1966). Stability parameters for comparing varieties. Crop Science. 6(1): 36-40. doi: 10.21 35/cropsci1966. 0011183X000600010011x

  6. Gauch, H.G. (1992). Statistical analysis for regional yiel trials: AMMI Analysis of Factorial designs. Elsevier, Amsterdam, The Netherlands.

  7. Gauch, H.G. (2006). Statistical analysis of yield trials by AMMI and GGE. Crop Science. 46: 1488-1500.

  8. Gauch, H.G. and Zobel, R.W. (1996). Optimal replication in selection experiments. Crop Science. 36(4): 838-843.

  9. Gurmu, F., Lire, E., Asfaw, A., Alemayehu, F., Rezene, Y. and Ambachew, D. (2012). GGE-biplot analysis of grain yield of faba bean genotypes in southern Ethiopia. Electronic Journal of Plant Breeding. 3(3): 898-907.

  10. Mahalingam, A., Manivannan, N., Kumar, K.B. (2019). Stability of promising greengram [Vigna radiata (L.) Wilczek] genotypes over seasons through AMMI analysis. Electronic Journal of Plant Breeding. 10(4): 1448-1453.

  11. Manivannan, N., Mahalingam, A., Bharathikumar, K. and Rajalakshmi, K. (2023). Stability analyses using regression and AMMI models for seed yield in blackgram. Legume Research: An International Journal. 46(2): 148-153. doi: 10.18805/LR-4403.

  12. Nandini, H., Parimala, K., Rukmini, K. (2022). Genetic analysis of diverse greengram (Vigna radiata L.) genotypes for combining ability and gene action for yield and its associated traits. Biological Forum-International Journal. 14(2): 344-348. 

  13. Naresh, T., Verma, S.K., Panwar, R.K., and Reddy, S.G.P.K., Rana, A., Bhatt, A., Mehra, A. and Pragati, K. (2025). AMMI biplot analysis of genotype × environment interactions in rainfed grown green gram [Vigna radiata (L.) Wilczek]. Legume Research-An International Journal. 48(12): 2032-2044. doi: 10.18805/LR-5374.

  14. Olivoto, T. and Lucio, A.D.C. (2020). metan: An R package for multi environment trial analysis. Methods in Ecology and Evolution. 11(6): 783-789.

  15. Piepho, H.P. (1996). Analysis of Genotype-by-Environment Interaction and Phenotypic Stability. In: Genotype by Environment Interaction. pp 151-174.

  16. Pillai, M.A., Anandhi, K. and Selvi, B. (2010). Stability analysis of yield in black gram.  Legume Research-An International Journal. 33(1): 70-71.

  17. Plant Breeding Tools (PBTools). Version: 1.3. (c) Copyright International Rice Research Institute (IRRI). 2013-2020. http://bbi.irri.org.

  18. Sanasam, S., Ali, S.M.B., Maibam, U., Kumawat, R. and Konjengbam, N.S. (2025). Evaluation of diverse greengram genotypes for seed yield stability under acidic soils by AMMI and GGE biplots analysis. Legume Research. 48(Special Issue): 44-51. doi: 10.18805/LR-5345.

  19. Shobanadevi, C., Elangaimannan, R. and Vadivel, K. (2023). Stability analysis for yield and its component characters in blackgram [Vigna mungo (L.) Hepper]. Indian Journal of Agricultural Research. 57(6): 737-739. doi: 10.18805/IJARe.A-5638.

  20. Sidramappa, Kuchanur, P.H., Shobharani, M., Arunkumar, B., Kulkarni, S.A. and Sowmya, H.C. (2024). Identification of stable and high yielding dual season genotypes in blackgram [Vigna mungo (L.) hepper]. Legume Research-An International Journal. 47(4): 519-524. doi: 10.18805/LR-4523.

  21. Yan, W. (2001). GGE Biplot: A windows application for graphical analysis of multi-environment trial data and other types of two-way data. Agronomy Journal. 93: 1111-1118.

  22. Yan, W. and Hunt, L.A. (2001). Interpretation of genotype × environment interaction for winter wheat yield in Ontario. Crop Science. 41(1): 19-25. doi:10. 2135/cropsci2001. 41119x.

  23. Yan, W. and Kang, M.S. (2003). GGE Bi-plot Analysis: A Graphical Tool for Breeders, Geneticists and Agronomists. Boca Raton FL: CRC Press.

  24. Yan, W. and Tinker, N.A. (2006). Biplot analysis of multi-environment trial data: Principles and applications. Canadian Journal of Plant Science. 86: 623-645. doi: 10.4141/P05-169.

  25. Yan, W., Hunt, L.A., Sheng, Q. and Szlavnics, Z. (2000). Cultivar evaluation and mega-environment investigation based on the GGE Bi-plot. Crop Science. 40: 597-605.

  26. Yihunie, T.A. and Gesesse, C.A. (2018). GGE Biplot analysis of genotype by environment interaction in field pea (Pisum sativum L.) genotypes in northwestern Ethiopia. Journal of Crop Science and Biotechnology. 21: 67-74.
Published In
Agricultural Science Digest
Article Metrics
Metrics Icon
48
Views
0
Citations
Reviewed By
Share Article
Download Article
In this Article
    Contact us
    Follow us

    Full Research Article

    Stability Analysis of Greengram [Vigna radiata (L.) Wilczek] Genotypes for Seed Yield in Telangana using Regression, AMMI, GGE and WAAS Biplot Models

    K
    K. Rukminidevi1,*
    0009-0004-6638-0920
    A
    A. Saritha1
    K
    K. Parimala1
    G
    G. Eswara Reddy1
    A
    A. Dinesh1
    N
    N. Lingaiah1
    B
    B. Supriya1
    1Agricultural Research Station, Professor Jayashankar Telangana Agricultural University, Madhira-507 203, Telangana, India.

    ABSTRACT

    Background: To increase area and production of greengram, breeders must develop high yielding, stable and adaptable varieties. This study aims to identify such superior greengram genotypes that perform consistently well across different locations.

    Methods: The experiment was carried out during Kharif, 2023 in six locations at different Agricultural Research Stations of Telangana state. A total of 15 genotypes were evaluated in RBD with three replications. The data on seed yield was subjected to statistical analysis. The G×E interaction was studied as per Eberhart and Russel model, AMMI, GGE biplot and WAAS model analysis.

    Result: Regression analysis revealed that genotypes MGG-385 and MGG-573 considered as stable. AMMI analysis of variance indicated that genotypes contributed 13.1% followed by environment (55.4%) and G×E interaction (20.48%). With further examination of GGE using AMMI analysis, two significant principal components were separated explaining 69.7% of variance interaction (PCI-42% and PC2-27.7%). The results of AMMI 1 and 2 biplot analysis, the genotypes MGG-385 and MGG-295 were considered as stable with high mean seed yield and recorded nearly zero IPCA1 score. The environments of Madhira and Tandur had recorded the lowest IPCA1 scores had less interaction effects. AMMI stability value and stability indexes were indicated MGG-385 was more stable. Y×WAAS graph revealed the genotypes VBN-4, MGG-570, MGG-571 were unstable had high seed yield, while the environments Palem, Tornala and Warangal provided high seed yield and presented good discrimination ability as had recorded high WAAS values. The genotypes MGG-573, MGG-556, MGG-385 and MGG-564 were broadly adopted. Based on all the stability models, the genotype MGG-385 was considered as stable with high mean seed yield across locations.

    INTRODUCTION

    Pulses are vital to india’s food and nutritional security offering a sustainable path for soil improvement and environmental health. Greengram [Vigna radiata (L.) Wilczek] plays a transformative role in diversifying Indian agriculture and addressing malnutrition in a largely vegetarian population. It holds significance as the third most important pulse crop in India. It provides an affordable source of nutrients, containing 24-26% protein, carbohydrates and essential minerals and vitamins. Furthermore, it plays a vital role in soil health through biological nitrogen fixation. As a centre pillar of the nations “pulses revolution”, mungbean has outperformed all other pulse crops, achieving a staggering growth rate of over 185% during the last one and half decade. Globally, Asia produces 90% of the mungbean crop, with India as the top producer, cultivating it over 3.7 million hectares with a production of 4.23 million tons. The development of short duration, photo thermo insensitive and disease resistant varieties has expanded their reach allowing them to thrive as Spring/Summer crops in north India and as vital relay crops in the rice fallows of the southern peninsula. (AICRP Kharif Pulses-Annual Report, 2025-26). This low and unstable production is primarily due to the inconsistent performance of varieties under varied environmental conditions. Therefore, a critical breeding goal, especially in the context of climate change adaptation, is to develop varieties that are less sensitive to variations in weather conditions to stabilize production (Sidramappa et al., 2024). An effective cultivar must possess a combination of both high yield and genetic stability (Yihunie and Gesesse, 2018). The challenge lies in the Genotype × Environment (G×E) interaction, which is often non-additive, meaning a genotype’s performance largely depends on the specific environment. Identifying cultivars with predictable performance requires a rigorous analysis of adaptability and stability (Pillai et al., 2010).  Stability analysis methods are necessary to recommend genotypes that consistently outperform and yield higher across various locations, seasons  and environments. Selecting robust and suitable methods for studying stability is a critical and challenging task for any breeder in a varietal development program (Piepho, 1996).
           
    Several models have been developed to dissect and address the G×E interaction. The Eberhart and Russell (1966) model suggested considering both linear (bi) and non-linear (S2di) components of the GxE interaction to determine a genotype’s phenotypic stability. The AMMI model is a hybrid analysis that uniquely separates main (additive) and interaction (multiplicative) effects (Gauch, 1992).  It uses principal component analysis (PCA) on the interaction portion to explain the G×E pattern in greater detail. AMMI also provides a visual representation via a biplot using the first Interaction Principal Component Axis (IPCA1) and mean yield to diagnose interaction patterns and identify stable genotypes (Gauch, 2006). The GGE biplot analysis is widely employed to detect and graphically indicate the GEI across multiple environmental trials (Yan et al., 2000). It is based on the first and second symmetrically scaled principal components derived from the Singular Value Decomposition of environment-centered data. GGE biplots are effective tools for facilitating variety evaluation and the delineation of mega-environments. WAAS is a novel technique that offers a superior framework by combining and improving upon AMMI and GGE (Olivoto and Lucio, 2020). Unlike AMMI, which often relies only on the first PC, WAAS incorporates all Principal Components (PCs) from the AMMI ANOVA, thus effectively capturing the entire variance of GEI. The WAAS biplot is used for the joint interpretation of a genotype’s average yield performance and stability across environments. To develop an effective selection strategy for future greengram breeding programs, this study used multivariate statistical techniques to identify stable and superior greengram genotypes, aiming to upscale cultivation and ensure consistent output.

    MATERIALS AND METHODS

    The experiment was conducted with 15 elite greengram genotypes during Kharif season, 2023 at six locations comprising of different Agroclimatic zones working under PJTAU, Telangana state, India (Table 1). The experimental material was planted in a RBD with three replications in each location adopting row to row and plant to plant spacing of 30 cm and 10 cm respectively. All the recommended package of practices and plant protection measures were followed to raise a good crop. The data on seed yield (kg/ha) was subjected to statistical analysis. The G×E interaction was studied i.e., Eberhart and Russel, (1966) analysis were carried out using the software P.B. tools (ver.1.3) developed by International Rice Research Institute Philippines. Additive main effects and multiplicative interaction (AMMI) (Gauch and Zobel, 1996), GGE biplot or site Regression model (Yan and Kang, 2003) and WAAS (Olivoto and Lucio, 2020) analyses were carried out using the software R package for multi environment trial analysis.

    Table 1: Details of greengram genotypes along with environments.

    RESULTS AND DISCUSSION

    Analysed data of each environment indicated that genotypes were found to be significant in all the six environments revealed greater magnitude of genetic variation for seed yield. Pooled analysis of variance for seed yield (Table 2) indicated that the genotypes, environments and genotype x environment were significant indicating the presence of variation among genotypes, environments and influence of G×E interaction.

    Table 2: Pooled analysis of variance of seed yield in greengram genotypes.


           
    Both linear and non linear components were found to be significant indicating the presence of both stable and non stable genotypes which was also confirmed by Manivannan et al. (2023). Stability parameters such as mean (xi), regression coefficient (bi) and deviation from regression (S2di) are considered to assess the stability of different genotypes for seed yield as suggested by Eberhart and Russel, (1966). The genotypes were categorized into three groups high yielders (939-1136 kg/ha), medium yield (827-880 kg/ha) and low yielders (742-811 kg/ha). Among the genotypes, MGG-556, MGG-570, MGG-564 and MGG-571 recorded high mean yield, regression coefficient more than unity (bi>1) and non significant S2di value indicating that these varieties perform better under good environment and were stable (Table 3). The genotype G9 (MGG-385), G4 (MGG-573) has recorded high yield, regression coefficient (bi=1) and non-significant S2di values, indicating these genotypes have an average response to environment and stable due to predictable nature and perform better across environments. The entries G13(MGG-295) and G6(MGG-576) and G1(MGG-563) recorded medium yield, bi=1 and non-significant S2di can perform better under poor environments and stable. The genotype G15 (VBN-4) has recorded bi value nearer to unity and significant S2di value indicating that this variety is suitable for good environment and unpredictable. The genotypes IPM-2-14, WGG-42, MGG-562 and MGG-565 were poor yielders and recorded non significant S2d and less regression coefficient (bi<1), hence seed yield did not improve with the improvement in the environment. However, due to the fact that the G×E conserts of both additive and multiplicative model, the recent researchers used the AMMI, GEE biplot and WAAS models. Hence the present data was used to analyze both models to compare the results.

    Table 3: Genotypic means with stability parametres for seed yield (kg/ha) in greengram genotypes.


           
    The analysis of variance for AMMI analysis elucidated that the significant difference existed among genotypes, environments and genotypes x environment interaction and witnessed the considerable influence of environments and interaction of genotypes with environments in expressing of the grain yield (Table 4). Further, genotypes contributed 13.18%, environments contribute (55.47%) more in total sum of squares followed by genotype x environment interaction (20.48%) revealing that genotypic architecture of the genotype, environment and their interaction in manifestation of the trait. The larger portion of the yield variation explained by environments indicated that the environments causing the most of the variation for seed yield. The seed yield over environments ranged from 588 kg/ha in Karimnagar (E1) to 1294 kg/ha in Tornala (E4). A large proportion of yield variation explained by environments indicated thet the environments were diverse with large difference among environments causing the most of the variation for grain yield. The genotypic seed yield ranged from 743 kg/ha G11(IPM 2-14) to 1136 kg/ha G7(MGG-556). G×E interaction was a crossover type with different yield ranking of genotypes across environments. The significant interaction in genotype and environment for yield validated the need to take more care while selecting the promising genotypes by considering stability and adaptability. Significant differences across locations were also earlier reported by Sanasam et al. (2025) using AMMI model. With further division of GGE using AMMI analysis, two significant principal components were separated explaining 69.7% of variance interaction (PCI 42% and PC2 27.7%). Earlier reports confirmed that in most of the cases the maximum genotype and environment interaction could be explained through using the first three PCA’s. Naresh et al. (2025); Sanasam et al. (2025) in greengram, Chandramohan et al. (2023) in rice evaluated the interactions between genotypes and environmental axes IPCA1 and IPCA2. While Mahalingam et al. (2019) reported IPCA1 alone may decide the G×E interaction in this study, whereas Manivannan et al. (2023) also reported that IPCA1, IPCA2 and IPAC3 contribute 46.3%, 37.5% and 16.0% towards interactions respectively.

    Table 4: AMMI and GGE biplot analysis of variance for seed yield in greengram genotypes.


           
    The results of the AMMI analysis further enlightened the relative contribution of the first two IPCA axes to the interactive effects by plotting with genotype and environment means as presented in Fig 1. In the biplot, environments are designated by the letter E followed by number 1 to 6 suffix (E1-E6) while, genotypes represented by numbers for 1 to 15 (G1-G15). When a variety and environment have the same sign on PCA1 axes their interaction is positive and if opposite their interaction is negative. Thus if a variety has a IPCAI score near to zero, it has small interaction effect and committed to the stable over wide environments. Conversely, varieties with high mean yield and large IPCA1 scores were considered as explicitly adapted to specific environments. Similar views were expressed in their studies by Manivannan et al., (2023); Shobanadevi et al., (2023); Naresh et al. (2025) and Sanasam et al. (2025).

    Fig 1: AMMI-I biplot mean seed yield (kg/ha) of the genotypes (G) and environment (E) against their respective IPCA1 (Y-axis) scores.


           
    The greengram genotypes G7(MGG-556) recorded high seed yield followed by G12 (MGG-570), G15 (VBN-4), G9(MGG 385) and G8 (MGG-571). In the present study G7(MGG556) has recorded highest seed yield value (horizontal axis) with positive IPCA1 score (vertical axis) of 9.91 followed by G12 (MGG-570) and G14 (MGG-564) indicating positive genotype × environmental interaction, hence these genotypes can be recommended for cultivation under favorable environments. The genotype G9(MGG-385) have recorded nearly zero IPCA1 score (0.03) (Table 5) indicating minimal interaction with the environment in which genotype showed above average yield and identified as most stable genotype with less yield variation across different environments. The other genotypes had less than the mean seed yield and found specific adoption to few tested environments. Most of the genotypic stability was attributed by environments E2 (Madhira) (-2.74) and E4 (Tandur)(-5.92) which had the lowest IPCA 1 scores had very little interaction effects and thus the performance of all genotypes in these environments was quite well. Hence these were considered as favorable environments for all the tested genotypes. With respect to GEI contribution E5 (Tornala) (17.84) and E3 (Palem) (13.99) exhibited the highest level of contribution as indicated by their high IPCA1 scores had the greatest effect on the G×E, so the comparative genotypes ranking in these environments were unstable.

    Table 5: Ranking of Genotypes according to yield and stability indexes.


           
    The environment having the long spokes is said to be more interactive (Fig 2) and among the environments E2 (Madhira) and E5 (Tornala) has the long spokes than E6 (Warangal), E4 (Tandur) and E3 (Palem) environments considering the genotypes. Those genotypes are nearer to origin are considered to be less interactive with environments. Based on this criterion genotypes G9 (MGG 385) within the orient and G13 (MGG295) close to the biplot origin had a smaller spoke of G´E were considered as less interactive genotypes with environments and generally stable in all environments. The genotypes G7(MGG-556), G12(MGG-570), G8(MGG-571) and G14(MGG-564) are far from the origin had the most variation due to environmental changes.

    Fig 2: AMMI-2 biplot (PCA1 on X-axis and PCA2 on Y-axis).


           
    Based on AMMI biplot 1 and 2, the genotype MGG-385 was identified as stable genotype with high seed yield. E4 (Tandur) and E3 (Palem) environments where the vectors were shortest and could discriminate between genotypes. In this graph, sites with short spokes did not exert strong interactive forces. Those with long spokes exerted strong interaction.
           
    To get a clear cut ranking of the genotypes regarding stability, AMMI stability value (ASV) stability indexes were computed (Table 5). G9 (MGG-385) showed the minimum ASV (2.117) value and ranked as number one regarding stability, while G8 (MGG 571) has showed highest ASV (21.83) value and showed poorest stability.
           
    GGE biplot helps in identifying GxE interaction pattern of data and provides us with clear picture of which genotype perform best in which environments and thus facilitates mega environment identification than AMMI (Table 4). Otherwise, both GGE and AMMI models are equivalent as far as their accuracy is considered (Gurmu et al., 2012). GGE biplot analysis also enables visual assessment of adaptability and stability. GGE biplot is presented with two principal components explaining a total of 75%. GGE variation (PC1=55.53%; PC2=19.9% represented in Fig 3. The first Principal component is represented on the X-axis and across its values is estimated yield i.e., genotypes that have higher PC1 values are considered to be more productive. The second principle component is represented on the Y-axis and presents the stability of the genotypes. Estimation of yield and stability of genotype was done by using so called AEC (average coordinates of the environment) method (Yan, 2001; Yan and Hunt, 2001). By projecting the genotypes on AEA axis, the genotypes are ranked by yield where the yield increases in the direction of arrow. Single arrowed lines in the Average Environment Coordination Abscissa (AEC) points to higher mean seed yield across environments (Fig 4). The double arrowed line in the AEC Coordinate representing highest variation in either direction. Hence, the genotypes G9 (MGG-385) and G13 (MGG-295) are stable, whereas others are highly interactive with environments. Hence, the genotypes G9 (MGG-385) and G13 (MGG-295) are found to be ideal with high mean and stability.

    Fig 4: Biplot of stability and mean performance of genotypes across environments.


           
    Discriminating and representativeness are the most important parameters of the GGE biplot when evaluating an environment. In Yan and Tinker, (2006) model a long environmental vector had high discriminating ability and short one had low discrimination therefore as shown in Fig 5 among tested locations E2(Madhira) and E5(Tornala) were identified as the potential environments for discriminating ability and representativeness. Yan et al. (2000); Laxuman et al., (2025) and Sanasam et al., (2025) also emphasized that the environments with long vectors are more discriminating and representativeness for consideration in future studies. Among the environmental vectors E3-E5, E2-E4, E2-E6 are found to be positively correlated (acute angle) E1-E6, E1-E2, E1-E3 vectors from obtuse angle and hence they are negatively correlated. E3-E6 form right angle and hence not correlated between them. Longer the distance between environmental vectors indicates that they are dissimilar in discriminating the genotypes. From the Fig 5, E2-E4 form one group whereas E1-E4 form two different groups. Environmental vector of E5 is the longest indicating that it is the environment with most discriminating ability followed by E2 (Madhira) and E3 (Palem) while E1 (Karimnagar) and E4 (Tandur) is least discriminating environments.

    Fig 5: GGE biplot –Discriminativeness and representativeness of genotypes for seed yield in greengram.


           
    What won where view of the GGE biplot is the best model for multi environment data for classifying the environments and also to select best performing genotypes in each (Yan et al., 2000). The equality line divides biplot into different sectors and winning genotypes are located on vertex of each sector. Genotypes located on the vertices of the polygon performs either the best or the poorest in one or the more environments. Vertex genotypes have longest vectors in their respective direction which is a measure of the responsiveness of the environment. The vertex genotypes G1 (MGG-563), G8 (MGG-571), G12 (MGG-570), G15 (VBN-4), G11 (IPM 2-14), G3 (MGG-565) and G5 (WGG-42) have longest vectors in their respective directions (Fig 6). The genotypes G9 (MGG-385) and G13 (MGG-295) considered as stable and performed better under all the environments. The genotypes G3 (MGG-565), G5 (WGG-42) and G11 (IPM 2-14) were best performers in mega environment consisting in E1(Karimnagar) while G1 (MGG-563), G8 (MGG-571) and G12 (MGG-570) perform well in mega environments consists of E5(Tornala), E3 (Palem) and E4 (Tandur) and G15(VBN-4) in E1 (Karimnagar) and E6 (Warangal) are the winning genotypes. The other genotypes G4 (MGG-573), G14 (MGG-564), G12 (MGG-570), G6(MGG-576) and G2(MGG-553) are poorest performers of all the environments because there is no environment in their sector. Hence for different mega environments, different genotypes have to be selected and deployed for each.

    Fig 6: What-won-Where GGE Biplot for yield (kg/ha).


           
    Y×WAAS biplot was divided into four quadrants which represented the classification of genotypes and environments (Fig 7). The quadrants in the graph represent (QI and QII) higher mean, (QIII and QIV) lower means. (Q1 and QIV) positive IPCA1 and (QII and QIII) negative IPCAI scores (Table 5). G3(MGG-565) and G1(MGG-563) grouped in the I quadrant were unstable genotypes with productivity below the grand mean and E1 environment was included with high discrimination ability as had high WAAS value. In quadrant II, G15 (VBN-4), G12 (MGG-570) and G8 (MGG-571) were present which were unstable genotypes, but had seed yield above the highest average value. The environments E2(Madhira), E3 (Palem), E5(Tornala) and E6(Warangal) included in the II quadrant provides high seed yield, they presented a good discrimination ability as had high WAAS values except E2 (Madhira). Genotypes G2 (MGG-553), G5 (WGG-42),G6 (MGG-576),G10 (MGG-562),G11 (IPM-2-14), G13 (MGG-295) included in the III quadrant had a low seed yield, low WAAS values so considered stable but no environment with high discrimination ability was identified. The genotypes G4 (MGG-573), G7 (MGG-556), G9 (MGG-385) and G14 (MGG-564) present in the IV quadrant were broadly adopted but their performance regarding yield were just nearer to mean. AMMI biplot found E2(Madhira) and E4 (Tandur) as favorable while E5 (Tornala) and E6 (Warangal) with the greatest GEI as unfavorable for testing. However, Y×WAAS grouped the environment only based on discriminating ability, so in this biplot E3 (Palem), E5 (Tornala) and E6 (Warangal) all grouped in II quadrant with good discriminating ability. Further, these environments have high WAAS values indicating that they have higher interactions with the studied genotypes. The higher mean performance of these environments regarding seed yield indicated that their influence on the yield was positive. Similar results regarding GEI from the WAAS biplot were also observed for chickpea genotypes by Arshad et al., (2024).

    Fig 7: Y×WAAS biplot for seed yield (kg/ha) on X-axis and weighted average of absolute scores on Y-axis.

    CONCLUSION

    Yield stability is significantly influenced by both environment and genotypes, with their major interaction effects defined by the first two components of the AMMI model. Two genotypes were identified as stable with above-average yield: G9 (MGG-385) (960 kg/ha), G13 (MGG-295) (880 kg/ha) and had nearly zero IPCA1 scores. The genotypic stability was largely driven by environments E2 (Madhira) and E4 (Tandur), which were favorable for high seed yield expression across most of the genotypes. While E5 (Tornala) recorded the highest mean seed yield (1294 kg/ha) and was the most ideal environment, E2 (Madhira) and E5 (Tornala) were the most representative. G9(MGG-385) specifically exhibited the best combination of high seed yield and stability across all environments. Multivariate models are thus beneficial for identifying stable genotypes in multi-location testing.

    ACKNOWLEDGEMENT

    It is a part of regular research work and authors are duly acknowledge PJTAU for providing financial and research facilities.
     
    Disclaimers
     
    The views and conclusions expressed in this article are solely those of the authors and do not necessarily represent the views of their affiliated institutions. The authors are responsible for the accuracy and completeness of the information provided, but do not accept any liability for any direct or indirect losses resulting from the use of the content.
     
    Informed consent
     
    Informed consent is not applicable as work is a crop based plant research.

    CONFLICT OF INTEREST

    The authors declare that there are no conflicts of interest.

    REFERENCES


    1. Annual Report-Vol-II (2025-26). ICAR-All India Coordinated Research Project on Kharif Pulses, IIPR, Khanpur.

    2. Arshad, W., Aleem, S., Khan, M.I., Saqib, M., Zeeshan, M., Mohsin, M.U., Malik, A., Sharif, I., Younas, A., Ihsan, F. and Tauseef, M. (2024). AMMI, BLUP and WAASB stability models for the selection of stable chickpea (Cicer arietinum L.) genotypes in different agroecological regions of Pakistan. Journal of Food and Agricultural Technology Research. 3(02): 66-81.

    3. Laxuman, Avinalappa, H., Kuchanur, P.H., Kumar, K.S. and Kumar, G.A. (2025). Seed yield stability assessment of chickpea genotypes through AMMI and Bi plot analysis. Legume Research-An International Journal. 48(2): 218-223. doi: 10.18805/LR-4911.

    4. Chandramohan, Y., Krishna, L., Srinivas, B., Rukmini, K., Sreedhar, S., Prasad, K.S., Kishore, N.S., Rani, C.V.D., Singh, T.V.J. and Jagadeeshwar. R. (2023). Stability analysis of short duration rice genotypes in Telangana using AMMI and GGE Bi-plot models. Environment Conservation Journal. 24(1): 243-252.

    5. Eberhart, S.A. and Russel, W.A. (1966). Stability parameters for comparing varieties. Crop Science. 6(1): 36-40. doi: 10.21 35/cropsci1966. 0011183X000600010011x

    6. Gauch, H.G. (1992). Statistical analysis for regional yiel trials: AMMI Analysis of Factorial designs. Elsevier, Amsterdam, The Netherlands.

    7. Gauch, H.G. (2006). Statistical analysis of yield trials by AMMI and GGE. Crop Science. 46: 1488-1500.

    8. Gauch, H.G. and Zobel, R.W. (1996). Optimal replication in selection experiments. Crop Science. 36(4): 838-843.

    9. Gurmu, F., Lire, E., Asfaw, A., Alemayehu, F., Rezene, Y. and Ambachew, D. (2012). GGE-biplot analysis of grain yield of faba bean genotypes in southern Ethiopia. Electronic Journal of Plant Breeding. 3(3): 898-907.

    10. Mahalingam, A., Manivannan, N., Kumar, K.B. (2019). Stability of promising greengram [Vigna radiata (L.) Wilczek] genotypes over seasons through AMMI analysis. Electronic Journal of Plant Breeding. 10(4): 1448-1453.

    11. Manivannan, N., Mahalingam, A., Bharathikumar, K. and Rajalakshmi, K. (2023). Stability analyses using regression and AMMI models for seed yield in blackgram. Legume Research: An International Journal. 46(2): 148-153. doi: 10.18805/LR-4403.

    12. Nandini, H., Parimala, K., Rukmini, K. (2022). Genetic analysis of diverse greengram (Vigna radiata L.) genotypes for combining ability and gene action for yield and its associated traits. Biological Forum-International Journal. 14(2): 344-348. 

    13. Naresh, T., Verma, S.K., Panwar, R.K., and Reddy, S.G.P.K., Rana, A., Bhatt, A., Mehra, A. and Pragati, K. (2025). AMMI biplot analysis of genotype × environment interactions in rainfed grown green gram [Vigna radiata (L.) Wilczek]. Legume Research-An International Journal. 48(12): 2032-2044. doi: 10.18805/LR-5374.

    14. Olivoto, T. and Lucio, A.D.C. (2020). metan: An R package for multi environment trial analysis. Methods in Ecology and Evolution. 11(6): 783-789.

    15. Piepho, H.P. (1996). Analysis of Genotype-by-Environment Interaction and Phenotypic Stability. In: Genotype by Environment Interaction. pp 151-174.

    16. Pillai, M.A., Anandhi, K. and Selvi, B. (2010). Stability analysis of yield in black gram.  Legume Research-An International Journal. 33(1): 70-71.

    17. Plant Breeding Tools (PBTools). Version: 1.3. (c) Copyright International Rice Research Institute (IRRI). 2013-2020. http://bbi.irri.org.

    18. Sanasam, S., Ali, S.M.B., Maibam, U., Kumawat, R. and Konjengbam, N.S. (2025). Evaluation of diverse greengram genotypes for seed yield stability under acidic soils by AMMI and GGE biplots analysis. Legume Research. 48(Special Issue): 44-51. doi: 10.18805/LR-5345.

    19. Shobanadevi, C., Elangaimannan, R. and Vadivel, K. (2023). Stability analysis for yield and its component characters in blackgram [Vigna mungo (L.) Hepper]. Indian Journal of Agricultural Research. 57(6): 737-739. doi: 10.18805/IJARe.A-5638.

    20. Sidramappa, Kuchanur, P.H., Shobharani, M., Arunkumar, B., Kulkarni, S.A. and Sowmya, H.C. (2024). Identification of stable and high yielding dual season genotypes in blackgram [Vigna mungo (L.) hepper]. Legume Research-An International Journal. 47(4): 519-524. doi: 10.18805/LR-4523.

    21. Yan, W. (2001). GGE Biplot: A windows application for graphical analysis of multi-environment trial data and other types of two-way data. Agronomy Journal. 93: 1111-1118.

    22. Yan, W. and Hunt, L.A. (2001). Interpretation of genotype × environment interaction for winter wheat yield in Ontario. Crop Science. 41(1): 19-25. doi:10. 2135/cropsci2001. 41119x.

    23. Yan, W. and Kang, M.S. (2003). GGE Bi-plot Analysis: A Graphical Tool for Breeders, Geneticists and Agronomists. Boca Raton FL: CRC Press.

    24. Yan, W. and Tinker, N.A. (2006). Biplot analysis of multi-environment trial data: Principles and applications. Canadian Journal of Plant Science. 86: 623-645. doi: 10.4141/P05-169.

    25. Yan, W., Hunt, L.A., Sheng, Q. and Szlavnics, Z. (2000). Cultivar evaluation and mega-environment investigation based on the GGE Bi-plot. Crop Science. 40: 597-605.

    26. Yihunie, T.A. and Gesesse, C.A. (2018). GGE Biplot analysis of genotype by environment interaction in field pea (Pisum sativum L.) genotypes in northwestern Ethiopia. Journal of Crop Science and Biotechnology. 21: 67-74.
    In this Article
      Contact us
      Follow us
      Published In
      Agricultural Science Digest

      Editorial Board

      View all (0)