# Genotype by environment interaction analysis for seed cotton yield stability under normal irrigation and drought stress conditions using numerical stability statistics

## TL;DR

This study identifies stable and high-yielding cotton genotypes under normal and drought conditions in Egypt using statistical analysis.

## Contribution

The paper compares parametric and non-parametric stability statistics for genotype selection under drought stress.

## Key findings

- Genotypes G5 and G8 showed stability with high and moderate yield under drought stress.
- AMMI analysis revealed significant genotype-environment interactions affecting seed cotton yield.
- Stability statistics were grouped into dynamic and static concepts based on correlations with yield.

## Abstract

One of the study’s main goals is to find high-yielding and stable genotypes in cotton under normal irrigation conditions (NIC) and drought stress conditions (DSC), as well as the comparison between parametric and non-parametric stability statistics. In order to achieve this objective, 24 cotton genotypes were evaluated under NIC and DSC during the 2019 and 2020 growing seasons (four environments) at the Sakha Agriculture Research Station in the Kafr El-Sheikh Governorate of Egypt. Every trial was set up using a randomized complete block design with three replications. According to the ANOVA, years under NIC and genotypes under DSC had a highly significant impact on the seed cotton yield. The AMMI analysis showed significant effects of environments (P < 0.01), genotypes, and their interaction (GEI) (P < 0.05) on seed cotton yield in four environments (two years and two irrigation conditions). The AMMI model successfully divided the variability by GEI into three principal component axes (PCs), only PC1 explained 87% of the total variability of GEI. When compared to NIC, seed cotton productivity across DSC was found to be significantly reduced, ranging from 8.54% (G5 genotype) to 42.38% (G3 genotype). The stability statistics have the ability to separate, rank, and detect stable genotypes in both irrigation conditions. According to Yi, YSi and TOP stability measures, the genotypes G1, G5, G20 and G19 were determined as the most stable genotypes (dynamically) with the highest yielding performance, while, the genotypes G8 and G4 were the most stable (statically) with moderate seed cotton yield using most other stability statistics. Based on rank, Spearman’s rank correlation, PCA biplot and heatmap data, the stability parameters can be divided into two major groups that corresponded to different dynamic and static concepts of stability under NIC and DSC. The first group includes YSi and TOP, which are strongly correlated with Yi (dynamic). While, the second group has other stability statistics, which negatively correlated with Yi, thus related to the static concept. Generally, the genotypes G5 and G8 were stable coupled with high and moderate yield, respectively. Therefore, these genotypes are used in the cotton breeding programs for the development of improved cotton varieties to address drought conditions in Egypt.

The online version contains supplementary material available at 10.1038/s41598-025-33327-6.