IBAS: Interaction-bridged association studies discovering novel genes underlying complex traits

Kernel-based framework building low-dimensional representations of pathway activity from transcriptomic reference data to weight SNPs and test gene-level association without explicitly enumerating interactions.

Genetic contributions to complex traits are largely mediated by gene-gene interaction networks, yet most association frameworks test only marginal single-gene effects. Direct interaction modeling remains blocked by combinatorial complexity and statistical instability. IBAS sidesteps this by incorporating pathway-level interaction patterns through low-dimensional representations of pathway activity derived from transcriptomic reference data, which guide SNP weighting and gene-level association testing within a kernel-based framework. In perturbation-based simulations IBAS is more stable and reproducible than conventional TWAS and gene-based methods, while keeping well-calibrated Type I error under phenotype permutation. Applied to the WTCCC datasets it recovers known genes and proposes novel candidates with modest marginal effects missed by standard approaches, with replication in an independent cohort and analyses across multiple reference tissues showing shared and tissue-specific signals; the code is available on GitHub.

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

The strength is calibration: proposing additional genes without degrading Type I error is exactly what separates a usable method from a false-positive generator. Two limits frame its reach: the WTCCC datasets are old and modest in size compared with today's biobanks, and reliance on transcriptomic reference data carries over the tissue-bias problem already familiar from TWAS. Above all, the "novel genes" remain statistical candidates without functional evidence: this is a hypothesis-generating tool, not an argument for classifying a variant in clinic.

Analysis by Dr Thibaut Benquey