A shifted multiplicative model cluster analysis for grouping environments without genotypic rank change

Area of Science:

• Agricultural science
• Biometrics
• Genetics

Background:

• Genotypic rank-change across trial sites complicates multi-location crop performance analysis.
• Identifying subsets of sites with stable genotypic performance is crucial for accurate breeding programs.

Purpose of the Study:

• To develop and validate a cluster method for identifying subsets of trial sites with negligible genotypic rank-change.
• To apply the shifted multiplicative model (SHMM) within a clustering framework for site grouping.

Main Methods:

• Utilized a cluster method based on the shifted multiplicative model (SHMM) and site regression model (SREG1).
• Defined site distance using residual sum of squares from SHMM1 or genotype sums of squares when rank-change occurred.
• Employed a dichotomous splitting procedure on a dendrogram to identify stable site clusters.

Main Results:

• Successfully identified five distinct groups of trial sites.
• The identified site subsets demonstrated adequate fit with SHMM1 and consistent primary site effects.
• The method effectively minimized genotypic rank-change interactions within the selected site groups.

Conclusions:

• The proposed cluster method using SHMM is effective for identifying subsets of trial sites with stable genotypic performance.
• This approach aids in reducing confounding effects of genotypic rank-change in multi-location trials.
• The findings support more reliable genetic analysis and breeding decisions.