The Journal of the American Statistical Association (JASA) has achieved a 2025 Impact Factor of 4.0, and is recognized, alongside JRSSB, Biometrika, and AOS, as one of the four top journals in statistics, holding significant academic influence in statistical theory and methods. Recently, the research findings of Professor ZHOU Xingcai's team from the School of Statistics and Data Science of Nanjing Audit University were published online in this prestigious journal. The paper is entitled “Fairness-Aware Gaussian Graphical Regression Models with Application to Brain Co-Expression QTL Studies”. Professor ZHOU Xingcai is the first author of the paper, with XU Zinan, a master's student of our university, as the second author. Professor Bei Jiang of the University of Alberta, Canada, serves as the corresponding author, and Professor Linglong Kong is a co-author. In February this year, Professor ZHOU Xingcai's team also published research on principal component analysis (PCA) in IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), a top journal in the field of artificial intelligence.
In biomedical big data research, genetic differences often exist among different populations (such as groups of different ethnicities or regions). When analyzing gene networks, traditional statistical models usually focus only on the network structure itself and tend to overlook fairness across subgroups; as a result, the genetic characteristics of certain groups may be masked, compromising the accuracy of research conclusions. To address this challenge, Professor ZHOU Xingcai's team proposed fairness-aware Gaussian graphical regression models (Fair RegGGMs). By introducing a fairness mechanism into the traditional model, the method enables the gene networks of different groups to be fairly represented while preserving the accuracy of statistical inference. The paper systematically establishes the statistical learning theory of Fair RegGGMs, validates the effectiveness of the method with glioblastoma multiforme (GBM) data, and uncovers gene regulatory relationships that traditional methods failed to detect, offering a new perspective for understanding subgroup-specific genetic regulatory mechanisms in GBM. The abstract and link to the full paper are provided below.
Xingcai Zhou, Zinan Xu, Bei Jiang & Linglong Kong (2026). Fairness-Aware Gaussian Graphical Regression Models with Application to Brain Co-Expression QTL Studies. Journal of the American Statistical Association.
Link: https://doi.org/10.1080/01621459.2026.2688572
