120th General Meeting of the KCS

Type Poster Presentation
Area Medicinal Chemistry
Room No. Exhibition Hall 2+3
Time 10월 19일 (목요일) 11:00~12:30
Code MEDI.P-331
Subject Development of QSAR models for predicting subchronic inhalation toxicity
Authors Do Hyung Kim, Byeong Hun Lee, Sung Kwang Lee*
Department of Chemistry, Hannam University, Korea
Abstract Subchronic inhalation toxicity is toxicological information obtained from long-term repeated exposure by the inhalation route. This toxicity data provides information on the hazards of chemicals during inhalation in the long term. However, there is currently no long-term inhalation toxicity information registered for a lot of chemicals. Because this test is complex experiments and it takes a long time and requires a lot of money. One solution to this problem is QSAR(Quantitative Structure-Activity Relationship). QSAR studies that satisfy regulatory requirements can provide useful tools for predicting long-term toxicity of untested chemicals. The aim of this study was to develop a Quantitative Structure-Activity Relationship (QSAR) model to predict sub-chronic toxicity inhalation. Subchronic inhalation toxicity data (rats, 13weeks, 5-7 days, 6hours) according to OECD test guideline 413 were obtained from echemportal website. Molecular descriptors and chemical fingerprint were calculated from 2D molecular structure using PreADMET and KNIME program. Data sets were divided into training set(60%) and test set(40%) based on structure diversity by RDKit fingerprint. 2D descriptors, fingerprint and machine learning methods such as multiple linear regression(MLR) and support vector machine(SVM) were used to develop predictive QSAR model. The chance correlation and predictability of these method were validated by y-scrambling method and external validation. The reliable prediction range of model is set in the kNN-based applicability domain(AD).
E-mail kdh19921121@naver.com