Multivariate phenotypic characterization for candidate parent selection in Indonesian tropical S5 maize (Zea mays) inbred lines

RIZKA AL FAIQOH (Department of Agronomy, Faculty of Agricultural Bio-Industry and Forestry, Universitas Brawijaya. Jl. Veteran, Malang 65145, East Java, Indonesia)
ERLANGGA DWI ATMOJO (Department of Agronomy, Faculty of Agricultural Bio-Industry and Forestry, Universitas Brawijaya. Jl. Veteran, Malang 65145, East Java, Indonesia)
DARMAWAN SAPTADI (Department of Agronomy, Faculty of Agricultural Bio-Industry and Forestry, Universitas Brawijaya. Jl. Veteran, Malang 65145, East Java, Indonesia)
ARIFIN NOOR SUGIHARTO (Department of Agronomy, Faculty of Agricultural Bio-Industry and Forestry, Universitas Brawijaya. Jl. Veteran, Malang 65145, East Java, Indonesia)

Biodiversitas Journal of Biological Diversity
Vol. 27 No. 9 (2026)

Abstract. Al Faiqoh R, Atmojo ED, Saptadi D, Sugiharto AN. 2026. Multivariate phenotypic characterization for candidate parent selection in Indonesian tropical S5 maize (Zea mays) inbred lines. Biodiversitas 27 (9): d270916. https://doi.org/10.13057/biodiv/d270916. Identifying suitable candidate parental lines is an essential first step for improving hybrid maize breeding efficiency. This study evaluated nine S5 maize inbred lines from a single breeding population maintained by CV. Blue Akari, Indonesia. The study used a Randomized Complete Block Design (RCBD) with three replications at one location and season in Batu City, East Java, Indonesia. Fourteen traits were analyzed using Analysis of Variance (ANOVA), Principal Component Analysis (PCA), Genotype × Trait (GT) biplot analysis, hierarchical clustering, and the Multi-trait Genotype-Ideotype Distance Index (MGIDI), with cluster and ranking stability verified through Euclidean-distance, leave-one-block-out, and trait-inclusion robustness checks. ANOVA detected significant genotype effects for ten of fourteen traits (p<0.05). PCA explained 72.05% of total variation (PC1 = 51.92%, PC2 = 20.13%), with PC1 associated mainly with plant architecture and ear-biomass traits and PC2 with flowering traits. Hierarchical clustering grouped the lines into three phenotypic clusters: G6 in the highest-biomass cluster, and G2 grouped with G1 and G3, though G2's separation from the high-biomass cluster proved more robust to methodological choice than the exact three-way partition. Under a nominal 15% selection intensity (a realized proportion of 11.1%), MGIDI identified G6 as closest to the predefined ideotype, achieving its stated objective for seven of ten traits, a ranking that remained stable when trait selection was varied but favored G4 marginally over G6 under an alternative, moderate-height ideotype. G6 and G2 are proposed as priority candidates for future crossing and combining-ability evaluation, with G4 as a secondary candidate, representing phenotypically rather than genetically complementary profiles. These findings provide a preliminary, single-environment framework for prioritizing candidate parental lines in early-stage hybrid maize breeding, pending multi-environment and testcross confirmation.