GGE Which-Won-Where Polygon
GGE which-won-where polygon for multi-environment trials online. Vertex genotypes, sectors and winners per sector.
When to use it
Use after a multi-environment trial (MET) to identify which genotype wins in which mega-environment via the Yan-Hunt (2000) polygon view of the GGE biplot. The vertex genotypes form a convex hull; perpendiculars from the origin to each polygon side partition environments into sectors.
When NOT to use it
Do not use without first checking that PC1 + PC2 explain a meaningful share of the G + GxE sum of squares (typically at least 60 percent); if not, the polygon view is misleading. Do not interpret a single MET season as a stable mega-environment partition; need multiple seasons.
What you get
GGE biplot with the convex hull drawn over vertex genotypes; perpendicular sector boundaries from the origin; the winner genotype per sector; cumulative percent of GxE explained by PC1 and PC2; environments labelled by sector membership.
Worked example
4 genotypes by 3 environments mean-yield matrix (t/ha). Same MET dataset is shared by all three GGE views.
G1: 5.2, 6.1, 4.8 G2: 5.5, 6.4, 5.1 G3: 5.8, 6.7, 5.4 G4: 6.1, 6.0, 5.7
Expected output: PC1 captures the bulk of G+GxE; convex hull identifies vertex genotypes; per-environment winners reported in the sector summary.
Source: StatVeda built-in example, paired with the analyse page sample for toolId 'ggepolygon'.
How to interpret the output
Each sector is a candidate mega-environment; the vertex genotype at each sector wins in those environments. Environments far from the origin and clustered in one sector confirm the mega-environment grouping. Vertex genotypes that appear in no sector are outliers; consider dropping them. Yan and Hunt (2000) introduced the polygon view; Yan and Kang (2003) is the textbook treatment.
Common pitfalls
- Reading the polygon when PC1+PC2 capture under 50 percent of GxE.
- Treating one season as a stable mega-environment definition.
- Confusing GGE polygon with AMMI biplot (different decomposition).
Try it in StatVeda
The GGE Which-Won-Where Polygon (Yan-Hunt 2000) engine runs entirely in the browser. No signup, no install, no data sent to a server. Paste your data, hit Run, copy the output.
Open GGE Which-Won-Where Polygon (Yan-Hunt 2000)References
- Yan, W. and Hunt, L. A. (2000). Cultivar evaluation and mega-environment investigation based on the GGE biplot. Crop Science, 40(3), 597 to 605.
- Yan, W. and Kang, M. S. (2003). GGE Biplot Analysis: A Graphical Tool for Breeders, Geneticists, and Agronomists. CRC Press, Boca Raton.