Gene Cluster Expression Index for IgA Nephropathy

Author(s): Aibing Rao, Weiyi Guo, Hong Cheng, Jun Xiao, Yan Zeng, Qiuyue Li, Jing Zhou.

from microarray and next-generation sequencing (NGS) have made possible for large-scale data analysis for biomarker discovery and pathogenesis. IgA nephropathy (IgAN) is the most popular nephritis in eastern Asia and in the world, but its pathogenesis has not been fully elucidated. We hereby study the gene cluster expression abnormality of IgAN compared to other kidney diseases and healthy controls using tissue microarray data. Methods: Publicly available needle biopsy tissue gene expression data sets for IgAN, other kidney diseases, and healthy control (HC) were used to develop the gene cluster expression analysis for nine representative pathways (clusters). By combining single variate prediction using ROC (receiver operating characteristic) and keyword search of gene function, nine gene clusters were heuristically determined. The gene expression status (up, down, normal) of a given gene was firstly determined by ROC, and then the gene cluster expression status, called the gene cluster expression index (GCEI), was determined by the percentages of abnormally expressed members. At last IgAN risks were assessed in the spectrum of GCEIs. Results: Samples were classified into normal (GCEI=0) or abnormal (GCEI=1) for nine predefined gene clusters respectively, and were classified into one of 10 combinatory cGCEI groups. The percentages of IgAN differed dramatically among the groups. The percentages in the abnormal groups (GCEI=1) ranged from 40% to 70% while the percentages were from 16% to 30% in the normal groups (GCEI=0) in both training and testing data sets. The percentages of IgAN corresponding to cGCEIs trended up when the number of abnormal clusters went from 0 to 9, starting from 8.06% to 88.24%. Conclusions: The binary GCEI is proposed to indicate whether a cluster of genes is normal (0) or abnormal (1) in terms of gene expression, and the categorical combinatory cGCEI represents the number of abnormal gene clusters. They are highly correlated to different IgAN risks so that they can be used as novel disease molecular sub-typing tools for future IgA nephropathy management and treatments.