Key points are not available for this paper at this time.
Clinical and biological insights from large-scale omics data are often limited by technical variability and analytical complexity. Here, we present BioinAI, a comprehensive framework that integrates an intelligent system with two algorithms, DeepAdvancer and stNiche, to enable effective data integration. Specifically, DeepAdvancer leverages a class-aware adversarial autoencoder to reconstruct gene expression profiles. When applied to 49,738 samples across 1,016 datasets, it uncovered a transcriptomic continuum and differential trajectory axes, which link diverse diseases through shared immune responses and distinct fate determinants. In the spatial context, stNiche leverages graph networks and symmetry-aware matching to identify functional cellular niches across heterogeneous slides. For instance, it identified a fibroblast-immune niche surrounding hair follicles in healthy skin that is lost in pathological states. BioinAI also provides an online conversational analysis platform, powered by multiple semi-agents, facilitating biological insight extraction from transcriptomic data.
Cui et al. (Fri,) studied this question.