Xuanjun Fang

Hainan Provincial Key Laboratory of Crop Molecular Breeding, Hainan Institute of Tropical Agricultural Resources (HITAR), Sanya, 572025, Hainan, China
Author    Correspondence author
Biological Evidence, 2026, Vol. 16, No. 3   
Received: 22 May, 2026    Accepted: 22 Jun., 2026    Published: 30 Jun., 2026

This is an open access article published under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Genome-wide association studies (GWAS) have identified a large number of loci associated with complex traits and diseases. However, most of these signals arise from linkage disequilibrium (LD) rather than directly reflecting causal variants, thereby limiting their mechanistic interpretability. Probabilistic fine-mapping addresses this limitation by introducing posterior inclusion probabilities (PIPs) and credible sets, shifting the inferential target from a single significant locus to a distribution of causal probabilities and enabling a systematic characterization of genetic uncertainty. In recent years, with the growing availability of functional annotation data, multi-ancestry studies, and multi-omics resources, fine-mapping methods have continued to expand in both model architecture and application scope. Nevertheless, a unified theoretical perspective across these methods remains lacking. In this study, we develop a unified statistical framework centered on causal configurations by systematically integrating fine-mapping and colocalization analyses within a Bayesian inferential framework. Under this framework, fastPAINTOR constructs annotation-informed priors using functional annotations, whereas CAVIAR and its extension MsCAVIAR strengthen likelihood-based constraints through LD structure and cross-study information. Colocalization analysis further extends the inferential target from a single-trait setting to a multi-trait space, enabling probabilistic modeling of cross-trait causal consistency. Accordingly, research on complex traits can be organized into a continuous inferential pipeline from GWAS to fine-mapping, and further to colocalization and transcriptome-wide association studies (TWAS), thereby progressively translating statistical associations into biological mechanistic interpretation. On this basis, we further propose a method-selection strategy based on inferential hierarchy, clarifying the trade-offs among computational complexity, causal resolution, and information sources across different methods, and summarizing a practical workflow of “hierarchical inference.” This framework is applicable not only to studies of human complex diseases, but also to applied contexts such as crop genetic improvement, where it can be used to assess causal consistency across environments and populations. By unifying fine-mapping and colocalization within the same causal inference layer, this study provides statistical genetics with a consistent conceptual language and analytical paradigm, thereby facilitating the systematic transition of complex trait research from association discovery to mechanistic interpretation and causal inference.

Complex traits; Fine-mapping; Causal inference; Colocalization analysis; Credible sets; Multi-omics integration