Prediction of Skin Sensitization with a Particle Swarm Optimized Support Vector Machine

被引:19
作者
Yuan, Hua [1 ,2 ,3 ]
Huang, Jianping [4 ]
Cao, Chenzhong [1 ,2 ,3 ]
机构
[1] Hunan Univ Sci & Technol, Minist Educ, Key Lab Theoret Chem & Mol Simulat, Xiangtan 411201, Peoples R China
[2] Hunan Prov Univ, Key Lab QSAR QSPR, Xiangtan 411201, Peoples R China
[3] Hunan Univ Sci & Technol, Sch Chem & Chem Engn, Xiangtan 411201, Peoples R China
[4] Zhejiang Univ, Coll Pharmaceut Sci, Pharmaceut Informat Inst, Hangzhou 310027, Peoples R China
基金
中国国家自然科学基金;
关键词
skin sensitization; guinea pig maximization test; murine local lymph node assay; support vector machine; particle swarm optimization; MECHANISTIC APPLICABILITY DOMAINS; CATEGORICAL QSAR MODELS; CLASSIFICATION; SELECTION; ALGORITHM;
D O I
10.3390/ijms10073237
中图分类号
Q5 [生物化学]; Q7 [分子生物学];
学科分类号
071010 ; 081704 ;
摘要
Skin sensitization is the most commonly reported occupational illness, causing much suffering to a wide range of people. Identification and labeling of environmental allergens is urgently required to protect people from skin sensitization. The guinea pig maximization test (GPMT) and murine local lymph node assay (LLNA) are the two most important in vivo models for identification of skin sensitizers. In order to reduce the number of animal tests, quantitative structure-activity relationships (QSARs) are strongly encouraged in the assessment of skin sensitization of chemicals. This paper has investigated the skin sensitization potential of 162 compounds with LLNA results and 92 compounds with GPMT results using a support vector machine. A particle swarm optimization algorithm was implemented for feature selection from a large number of molecular descriptors calculated by Dragon. For the LLNA data set, the classification accuracies are 95.37% and 88.89% for the training and the test sets, respectively. For the GPMT data set, the classification accuracies are 91.80% and 90.32% for the training and the test sets, respectively. The classification performances were greatly improved compared to those reported in the literature, indicating that the support vector machine optimized by particle swarm in this paper is competent for the identification of skin sensitizers.
引用
收藏
页码:3237 / 3254
页数:18
相关论文
共 31 条
[1]  
Al kazemi B., 2002, 4 INT WORKSH FRONT E
[2]   Allergic contact dermatitis [J].
Andersen, KE ;
Frankild, S .
CLINICS IN DERMATOLOGY, 1997, 15 (04) :645-654
[3]  
[Anonymous], 2011, Pei. data mining concepts and techniques
[4]   Skin sensitization: Reaction mechanistic applicability domains for structure-activity relationships [J].
Aptula, AO ;
Patlewicz, G ;
Roberts, DW .
CHEMICAL RESEARCH IN TOXICOLOGY, 2005, 18 (09) :1420-1426
[5]   Mechanistic applicability domains for nonanimal-based prediction of toxicological end points: General principles and application to reactive toxicity [J].
Aptula, Aynur O. ;
Roberts, David W. .
CHEMICAL RESEARCH IN TOXICOLOGY, 2006, 19 (08) :1097-1105
[6]  
Chang C.-C., LIBSVM: a Library for Support Vector Machines
[7]  
*DHHS, 2000, DHHS NIOSH PUBL
[8]  
*ECETOC, 2004, WORKSH REG ACC Q SAR
[9]   Computer-aided knowledge generation for understanding skin sensitization mechanisms: The TOPS-MODE approach [J].
Estrada, E ;
Patlewicz, G ;
Chamberlain, M ;
Basketter, D ;
Larbey, S .
CHEMICAL RESEARCH IN TOXICOLOGY, 2003, 16 (10) :1226-1235
[10]   Structure-activity models for contact sensitization [J].
Fedorowicz, A ;
Singh, H ;
Soderholm, S ;
Demchuk, E .
CHEMICAL RESEARCH IN TOXICOLOGY, 2005, 18 (06) :954-969