ASSOCIATION OF FEATURE GENE EXPRESSION WITH STRUCTURAL FINGERPRINTS OF CHEMICAL COMPOUNDS

被引:4
作者
Li, Yun [1 ,2 ]
Tu, Kang [1 ]
Zheng, Siyuan [1 ]
Wang, Jingfang [3 ]
Li, Yixue [1 ,3 ,4 ]
Hao, Pei [1 ,4 ]
Li, Xuan [1 ]
机构
[1] Chinese Acad Sci, Shanghai Inst Biol Sci, Key Lab Syst Biol, Key Lab Synthet Biol, Shanghai 200031, Peoples R China
[2] Chinese Acad Sci, Grad Sch, Shanghai 200031, Peoples R China
[3] Shanghai Jiao Tong Univ, Shanghai Ctr Syst Biomed, Shanghai 200240, Peoples R China
[4] Shanghai Ctr Bioinformat Technol, Shanghai 200235, Peoples R China
基金
中国国家自然科学基金; 国家高技术研究发展计划(863计划);
关键词
Feature genes; machine learning; chemical structure; similarity; PREDICTION; INHIBITORS; TOXICITY; SYSTEMS;
D O I
10.1142/S0219720011005446
中图分类号
Q5 [生物化学];
学科分类号
071010 ; 081704 ;
摘要
Exploring the relationship between a chemical structure and its biological function is of great importance for drug discovery. For understanding the mechanisms of drug action, researchers traditionally focused on the molecular structures in the context of interactions with targets. The newly emerged high-throughput "omics" technology opened a new dimension to study the structure-function relationship of chemicals. Previous studies made attempts to introduce transcriptomics data into chemical function investigation. But little effort has been made to link structural fingerprints of compounds with defined intracellular functions, i.e. expression of particular genes and altered pathways. By integrating the chemical structural information with the gene expression profiles of chemical-treated cells, we developed a novel method to associate the structural difference between compounds with the expression of a definite set of genes, which were called feature genes. A subtraction protocol was designed to extract a minimum gene set related to chemical structural features, which can be utilized in practice as markers for drug screening. Case studies demonstrated that our approach is capable of finding feature genes associated with chemical structural fingerprints.
引用
收藏
页码:503 / 519
页数:17
相关论文
共 26 条
[1]  
Brown A C, 1868, J Anat Physiol, V2, P224
[2]   NEAREST NEIGHBOR PATTERN CLASSIFICATION [J].
COVER, TM ;
HART, PE .
IEEE TRANSACTIONS ON INFORMATION THEORY, 1967, 13 (01) :21-+
[3]  
Daylight Information Systems, 2008, DAYLIGHT THEORY FING
[4]   In silico prediction of drug toxicity [J].
Dearden, JC .
JOURNAL OF COMPUTER-AIDED MOLECULAR DESIGN, 2003, 17 (02) :119-127
[5]   Reoptimization of MDL keys for use in drug discovery [J].
Durant, JL ;
Leland, BA ;
Henry, DR ;
Nourse, JG .
JOURNAL OF CHEMICAL INFORMATION AND COMPUTER SCIENCES, 2002, 42 (06) :1273-1280
[6]   Progress in predicting human ADME parameters in silico [J].
Ekins, S ;
Waller, CL ;
Swaan, PW ;
Cruciani, G ;
Wrighton, SA ;
Wikel, JH .
JOURNAL OF PHARMACOLOGICAL AND TOXICOLOGICAL METHODS, 2000, 44 (01) :251-272
[7]   Computer systems for the prediction of toxicity: an update [J].
Greene, N .
ADVANCED DRUG DELIVERY REVIEWS, 2002, 54 (03) :417-431
[8]   Blue Obelisk - Interoperability in chemical informatics [J].
Guha, Rajarshi ;
Howard, Michael T. ;
Hutchison, Geoffrey R. ;
Murray-Rust, Peter ;
Rzepa, Henry ;
Steinbeck, Christoph ;
Wegner, Jorg ;
Willighagen, Egon L. .
JOURNAL OF CHEMICAL INFORMATION AND MODELING, 2006, 46 (03) :991-998
[9]  
Hall M., 2009, SIGKDD Explorations, V11, P10, DOI DOI 10.1145/1656274.1656278
[10]  
Hall M. A., 1998, Correlation-based feature subset selection for machine learning