Twenty years of ASCA: a systematic review of applications for ANOVA-simultaneous component analysis

BACKGROUND: ANOVA-simultaneous component analysis (ASCA) integrates ANOVA-based variance decomposition with latent variable modelling to quantify, validate and interpret factor-related variation in high-dimensional analytical signals. Over the past two decades, ASCA has expanded through methodological variants (e.g., ASCA+ and mixed-model extensions) and has been adopted across analytical chemistry, agrifood science, biology and metabolomics. Despite this growth, practical use and reporting remain heterogeneous. The key problem is the lack of a consolidated, evidence-based picture of how ASCA is actually implemented and interpreted in the applied literature. RESULTS: Following PRISMA guidelines, we systematically screened the literature and identified 158 peer-reviewed ASCA application studies published since the original formulation of the method. Each article was annotated for application domain, data type and analytical technique, experimental design complexity, factorisation choices, validation strategies (including permutation testing), treatment of residuals, and the way component models and effect sizes were reported. The corpus is dominated by agrifood, biological and metabolomics applications, and most studies involve two or more factors, often including a time component that does not necessarily imply a longitudinal design. Across domains, we observe substantial inconsistency in how unbalanced designs and interactions are encoded, how effect magnitudes are quantified and normalised, how permutation schemes are specified and justified, and whether residual structure is examined to support conclusions or to detect confounding variation. SIGNIFICANCE: This review maps where and how ASCA is used in real experiments and pinpoints recurring pitfalls that limit transparency and reproducibility. By synthesising current practice into recommendations for design specification, effect-size reporting, validation and residual interpretation, it provides a practical basis for more robust and comparable ASCA studies and helps guide future applications and methodological development.