Laboratory of Nanoscale Characterization & Environmental Chemistry, Department of Chemistry, College of Natural Sciences, Hanyang University, Seoul
As the number of nano-products has been sharply increasing in recent years, there are more and more public concerns about the safety of nanomaterial and related products. Generally, toxicity of nanoparticles is evaluated through in vitro (experiments with cultured cells) and in vivo (experiments with living organisms) protocols which require much time and labor cost.
With the development of in silico approach (building and applying computational models for the toxicity prediction of nanomaterials), time and labor cost for the hazard assessment could be reduced. However, quality of published data on nanomaterials has not been paid much attention during process of building Quantitative Structure-Activity Relationship (QSAR) models.
In this study, we studied the effect of published data quality on the performance of QSAR models by using a set of scoring rules based on their physico-chemical (PChem) properties. The data sets were built from 122 published journal articles on metallic nanoparticles cytotoxicity and two algorithms, such as Support Vector Machines and Random Forests, were used for the performance evaluation of PChem score based screening.
As data quantity has been increasing exponentially, our study provided a useful tool for QSAR data quality assessment and choosing suitable datasets for QSAR modeling.
ABOUT THE PRESENTER
Xuan-Tung Trinh is a Ms/PhD student at Nano Characterization & Environmental Chemistry Laboratory, Department of Chemistry, College of Natural Science, Hanyang University, Seoul, South Korea since January 2016.
Xuan-Tung is studying about Quantitative Nano-structure Toxicity Relationship (QNTR) for risk assessment of nanomaterials. He is interested in interaction between nanomaterials and cells, data analysis and machine learning.
His recent activities include nanoparticle characterization using analytical instruments such as UV-Vis spectroscopy, Dynamic Light Scattering, Inductive Couple Plasma Mass Spectrometry, synchrotron soft X-ray Microscopy, collecting cellular toxicity data of nanomaterials from experiments and literatures and applying machine learning techniques for making prediction models of nanomaterials toxicity
|1||Ray, Paresh Chandra, Hongtao Yu, and Peter P. Fu. Journal of environmental science and health. Part C, Environmental carcinogenesis & ecotoxicology reviews, 27(1), 1-35, (2009).|
|2||Oh, E., Liu, R., Nel, A., Gemill, K. B., Bilal, M., Cohen, Y., & Medintz, I. L. Nature Nanotechnology, (2015).|
|3||Lubinski, L.et al., SAR and QSAR in Environmental Research, 24, 995–1008, (2013).|