Pigmentation has long served as a powerful system for exploring gene-trait relationships, yet much of the field has focused on a relatively narrow group of well-established genes involved in melanin production and pigment cell differentiation. Recent advances, however, have allowed pigmentation to be studied through a more comprehensive framework. By combining artificial intelligence (AI)-driven phenotyping with genomic mapping approaches such as genome-wide association studies, QTL mapping, and structural variant analysis, a broader range of pigmentation regulators has been identified across diverse animal taxa. This review highlights studies where AI methods, including deep learning, self-supervised modeling, and pattern recognition, have been used to quantify complex pigmentation traits in animals. These approaches have enabled the discovery of non-classical pigmentation genes involved in membrane trafficking, intracellular signaling, structural organization, and non-coding regulation. Rather than displacing the classical pigmentation paradigm, these findings extend it, revealing a wider set of genetic contributors to coloration and pattern diversity. We introduce the term AI-pigmentomics to describe the integration of AI-driven phenotyping with genomic mapping, as part of the broader emergence of AI-omics. Together, AI and genomic mapping are reshaping our understanding of pigmentation by uncovering unexpected biological mechanisms and providing a framework for investigating pigmentation in both model and non-model species.
Ahi et al. (Thu,) studied this question.