Standardized in vitro phenotypes provide genetically tractable traits that integrate GWAS, QTL mapping, and multi-omics, enabling mechanistic insight and genetics-informed improvement of plant regeneration. Plant tissue culture (PTC) has traditionally been employed as a technical tool for propagation, transformation, and germplasm conservation. This review reframes in vitro culture systems as quantitative phenotyping platforms that enable rigorous genetic dissection of complex traits, such as regeneration competence, stress tolerance, and developmental plasticity. By operating under highly standardized and reproducible conditions, in vitro systems reduce environmental noise and expose genotype-dependent variation that is often obscured in field or greenhouse experiments. When formalized as quantitative traits, culture responses, including callus induction, organogenesis, somatic embryogenesis, and stress survival, become amenable to genome-wide association studies (GWAS) and quantitative trait locus (QTL) mapping. Synthesizing evidence across diverse plant species, this review demonstrates that in vitro traits are genetically structured, polygenic, and biologically meaningful, reflecting coordinated regulation of hormonal signaling, chromatin accessibility, stress response pathways, and developmental gene networks. Numerous GWAS and QTL studies have identified loci and candidate genes underlying regeneration efficiency, embryogenic competence, and stress-responsive culture performance, confirming that tissue culture responsiveness is rooted in genomic architecture rather than protocol effects alone. The interpretive power of genetic mapping is substantially enhanced through integration with bioinformatic and multi-omics layers, including transcriptomics, epigenomics, proteomics, metabolomics, and network-based analyses, thereby refining candidate prioritization and facilitating mechanistic inference. This review further examines the dual role of somaclonal variation as both a constraint on clonal fidelity and a source of genetic and epigenetic diversity that can reveal cryptic genotype-phenotype relationships. Emphasis is placed on study design principles, phenotyping challenges, statistical modeling strategies, and integrative pipelines that connect genotype to in vitro phenotype and ultimately to breeding and biotechnology applications. This synthesis highlights in vitro culture systems as powerful, genetics-informed platforms for precision breeding, gene validation, and supporting efforts to more predictably and rationally manipulate complex traits, particularly in horticultural and clonally propagated crops.

Maize ear architecture is a critical determinant of grain yield, yet its genetic basis remains incompletely understood. This study aims to explore novel genetic variations generated by radiation-induced mutagenesis, systematically dissect the genetic basis of key maize ear traits and develop molecular markers capable of effectively distinguishing germplasm with contrasting ear phenotypes. We conducted an integrated genomic analysis of four key ear traits (ear length (EL), diameter (ED), ear row number (ERN), and kernel number per row (KNR)) using two radiation-mutagenized populations: 1355 multi-background radiation-induced mutant lines (RMLs) derived from three elite inbred lines (ZH6218, T32, and ZNC442), and 626 radiation-induced mutant lines derived from the ZH6218 background (ZF6218), grown in two environments. Phenotypic evaluation revealed substantial variation, with ED and ERN showing high heritability, while KNR exhibited stronger genotype-environment interactions. Genome-wide association studies (GWAS) identified 53 suggestive QTNs, including three pleiotropic loci associated with both ED and KNR. Linkage mapping in ZF6218 uncovered 36 segregation distortion loci (SDLs), with eight novel loci contributing to ear morphology. Notably, a prominent KNR-associated SDL (SDL-KNR8.2) co-localized with a GWAS peak (S08_18815086), narrowing the candidate interval to 575 kb on chromosome 8 harboring 22 genes. Spatiotemporal expression profiling suggested three distinct gene clusters: early apical, intermediate elongation, and late differentiation, implicating hierarchical control of meristem development. In silico candidate gene analysis identified a C/T SNP in the second exon of Zm00001d008757 that is significantly associated with KNR variation, which may serve as a candidate marker for breeding selection. These findings provide a theoretical basis and candidate resources for understanding maize ear development, and warrant further validation before application in molecular breeding.