Peptidome data-driven comprehensive individualized monitoring of membranous nephropathy with machine learning

作     者：Zixing Xu Ruiying Chen Chuanming Hao Qionghong Xie Chunhui Deng Nianrong Sun Zixing Xu;Ruiying Chen;Chuanming Hao;Qionghong Xie;Chunhui Deng;Nianrong Sun

作者机构：Department of Gastroenterology and HepatologyZhongshan Hospitaland Department of ChemistryFudan UniversityShanghai 200433China Division of NephrologyHuashan HospitalFudan UniversityShanghai 200040China School of Chemistry and Chemical EngineeringNanchang UniversityNanchang 330031China 

出 版 物：《Chinese Chemical Letters》 (中国化学快报（英文版）)

年 卷 期：2024年第35卷第5期

页      面：402-406页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 1002[医学-临床医学] 081104[工学-模式识别与智能系统] 100210[医学-外科学(含：普外、骨外、泌尿外、胸心外、神外、整形、烧伤、野战外)] 08[工学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 10[医学] 

基　　金：financially supported by National Key R&D Program of China(No.2018YFA0507501) the National Natural Science Foundation of China(Nos.22074019,21425518,22004017) Shanghai Sailing Program(No.20YF1405300) 

主　　题：Membranous nephropathy Serum peptidome Machine learning Disease diagnosis 

摘      要：As the most common pathological type of nephrotic syndrome,membranous nephropathy(MN)presents diversity in progression trends,facing severe *** precise discrimination of MN from healthy people,other types of nephrotic syndrome or those with therapeutic remission has always been huge challenge in clinics,not to mention comprehensive individualized monitoring relied on minimally invasive molecular detection ***,we construct a functionalized pore architecture to couple with machine learning to aid all-round peptidome enrichment and data profiling from hundreds of human serum samples,and finally establish a set of defined peptide panel consisting of 12 specific feature *** addition to the realization of above-mentioned precise discrimination with more than 97%of sensitivity,88%of accuracy and f1 score,the simultaneously comprehensive individualized monitoring for MN can also be achieved,including conventionally screening diagnosis,congeneric distinction and prognostic *** work greatly advances the development of peptidome data-driven individualized monitoring means for complex diseases and undoubtedly inspire more devotion into molecular detection field.