TY - JOUR
T1 - Interpretable prediction of non-genotoxic hepatocarcinogenic chemicals
AU - Tung, Chun Wei
AU - Jheng, Jhao Liang
N1 - Funding Information:
The authors would like to acknowledge the financial support from National Science Council of Taiwan ( NSC 101-2311-B-037-001-MY2 ), Kaohsiung Medical University Research Foundation ( KMU-M103009 ), NSYSU-KMU Joint Research Project ( NSYSUKMU103-P002 ) and National Health Research Institutes ( EH-103-PP-09 ). We especially thanks Dr. Chia-Chi Wang, who is one of the faculty of Kaohsiung Medical University, for her constructive suggestions and comments.
PY - 2014/12/5
Y1 - 2014/12/5
N2 - The assessment of non-genotoxic hepatocarcinogenicity of chemicals relies on time-consuming rodent bioassays. The development of alternative methods for non-genotoxic hepatocarcinogenicity could help the identification of potential hepatocarcinogenic chemicals. This study evaluated four types of features for the interpretable prediction of non-genotoxic hepatocarcinogenic chemicals including chemical-chemical interactions (CCI), chemical-protein interactions (CPI), chemical descriptors (QSAR) and gene expression profiles (TGx). Based on the results of decision tree classifiers, the CPI-based features perform best with independent test accuracies of 90% and 86% for interaction scores from combined scores and databases, respectively. Informative features were identified and analyzed to give insights into the non-genotoxic hepatocarcinogenicity of chemicals. The difference between CPI scores and gene expression profiles for the identified important proteins shows that CPI could play more important roles in non-genotoxic hepatocarcinogenicity.
AB - The assessment of non-genotoxic hepatocarcinogenicity of chemicals relies on time-consuming rodent bioassays. The development of alternative methods for non-genotoxic hepatocarcinogenicity could help the identification of potential hepatocarcinogenic chemicals. This study evaluated four types of features for the interpretable prediction of non-genotoxic hepatocarcinogenic chemicals including chemical-chemical interactions (CCI), chemical-protein interactions (CPI), chemical descriptors (QSAR) and gene expression profiles (TGx). Based on the results of decision tree classifiers, the CPI-based features perform best with independent test accuracies of 90% and 86% for interaction scores from combined scores and databases, respectively. Informative features were identified and analyzed to give insights into the non-genotoxic hepatocarcinogenicity of chemicals. The difference between CPI scores and gene expression profiles for the identified important proteins shows that CPI could play more important roles in non-genotoxic hepatocarcinogenicity.
KW - Chemical-chemical interaction
KW - Chemical-protein interaction
KW - Decision tree
KW - Non-genotoxic hepatocarcinogenicity
KW - Quantitative structure-activity relationship
KW - Toxicogenomics
UR - http://www.scopus.com/inward/record.url?scp=84906935725&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84906935725&partnerID=8YFLogxK
U2 - 10.1016/j.neucom.2014.05.073
DO - 10.1016/j.neucom.2014.05.073
M3 - Article
AN - SCOPUS:84906935725
VL - 145
SP - 68
EP - 74
JO - Neurocomputing
JF - Neurocomputing
SN - 0925-2312
ER -