Prediction of developmental chemical toxicity based on gene networks of human embryonic stem cells

Predictive toxicology using stem cells or their derived tissues has gained increasing importance in biomedical and pharmaceutical research. Here, we show that toxicity category prediction by support vector machines (SVMs), which uses qRT-PCR data from 20 categorized chemicals based on a human embryonic stem cell (hESC) system, is improved by the adoption of gene networks, in which network edge weights are added as feature vectors when noisy qRT-PCR data fail to make accurate predictions. The accuracies of our system were 97.5–100% for three toxicity categories: neurotoxins (NTs), genotoxic carcinogens (GCs) and non-genotoxic carcinogens (NGCs). For two uncategorized chemicals, bisphenol-A and permethrin, our system yielded reasonable results: bisphenol-A was categorized as an NGC, and permethrin was categorized as an NT; both predictions were supported by recently published papers. Our study has two important features: (i) as the first study to employ gene networks without using conventional quantitative structure-activity relationships (QSARs) as input data for SVMs to analyze toxicogenomics data in an hESC validation system, it uses additional information of gene-to-gene interactions to significantly increase prediction accuracies for noisy gene expression data; and (ii) using only undifferentiated hESCs, our study has considerable potential to predict late-onset chemical toxicities, including abnormalities that occur during embryonic development.

[1]  Edward J. Perkins,et al.  In vitro gene regulatory networks predict in vivo function of liver , 2010, BMC Systems Biology.

[2]  J. Benzecri.  L'analyse des données@@@L'analyse des donnees , 1975 .

[3]  T. Poggio,et al.  Multiclass cancer diagnosis using tumor gene expression signatures , 2001, Proceedings of the National Academy of Sciences of the United States of America.

[4]  T. Yamanaka,et al.  The TAO-Gen Algorithm for Identifying Gene Interaction Networks with Application to SOS Repair in E. coli , 2004, Environmental health perspectives.

[5]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[6]  Jeroen L A Pennings,et al.  Transcriptomic concentration-response evaluation of valproic acid, cyproconazole, and hexaconazole in the neural embryonic stem cell test (ESTn). , 2012, Toxicological sciences : an official journal of the Society of Toxicology.

[7]  Y Sakuratani,et al.  Hazard Evaluation Support System (HESS) for predicting repeated dose toxicity using toxicological categories , 2013, SAR and QSAR in environmental research.

[8]  H. Toyoshiba,et al.  Multi-Parametric Profiling Network Based on Gene Expression and Phenotype Data: A Novel Approach to Developmental Neurotoxicity Testing , 2011, International journal of molecular sciences.

[9]  Pierre R. Bushel,et al.  CEBS—Chemical Effects in Biological Systems: a public data repository integrating study design and toxicity data with microarray and proteomics data , 2007, Nucleic Acids Res..

[10]  Raffaella Corvi,et al.  The carcinoGENOMICS project: critical selection of model compounds for the development of omics-based in vitro carcinogenicity screening assays. , 2008, Mutation research.

[11]  J. Do.  Neurotoxin-Induced Pathway Perturbation in Human Neuroblastoma SH-EP Cells , 2014, Molecules and cells.

[12]  Jan G Hengstler,et al.  Compound selection for in vitro modeling of developmental neurotoxicity. , 2012, Frontiers in bioscience.

[13]  J. Kleinjans,et al.  A transcriptomics-based in vitro assay for predicting chemical genotoxicity in vivo. , 2012, Carcinogenesis.

[14]  Rafael A. Irizarry,et al.  Bioinformatics and Computational Biology Solutions using R and Bioconductor , 2005 .

[15]  Anna Beronius,et al.  The influence of study design and sex-differences on results from developmental neurotoxicity studies of bisphenol A: implications for toxicity testing. , 2013, Toxicology.

[16]  Marta Hoffmann,et al.  Bisphenol A induce ovarian cancer cell migration via the MAPK and PI3K/Akt signalling pathways. , 2014, Toxicology letters.

[17]  Shuk-Mei Ho,et al.  4 Epigenetically Regulates Phosphodiesterase Type 4 Variant Increases Susceptibility to Prostate Carcinogenesis and Developmental Exposure to Estradiol and Bisphenol A , 2006 .

[18]  T W Schultz,et al.  Structure-toxicity relationships for benzenes evaluated with Tetrahymena pyriformis. , 1999, Chemical research in toxicology.

[19]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[20]  Gordon K. Smyth,et al.  limma: Linear Models for Microarray Data , 2005 .

[21]  H. Sone,et al.  Prenatal exposure to permethrin influences vascular development of fetal brain and adult behavior in mice offspring , 2013, Environmental toxicology.

[22]  H. Yamada,et al.  Evaluation of DNA microarray results in the Toxicogenomics Project (TGP) consortium in Japan. , 2012, The Journal of toxicological sciences.

[23]  Tsuyoshi Kato,et al.  Classification of heterogeneous microarray data by maximum entropy kernel , 2007, BMC Bioinformatics.

[24]  M. Tsuda,et al.  Repression of activity-dependent c-fos and brain-derived neurotrophic factor mRNA expression by pyrethroid insecticides accompanying a decrease in Ca(2+) influx into neurons. , 2000, The Journal of pharmacology and experimental therapeutics.

[25]  Isabelle Guyon,et al.  An Introduction to Variable and Feature Selection , 2003, J. Mach. Learn. Res..

[26]  Bo-Han Su,et al.  Predictive Toxicology Modeling: Protocols for Exploring hERG Classification and Tetrahymena pyriformis End Point Predictions , 2012, J. Chem. Inf. Model..

[27]  Jean Armengaud,et al.  High-throughput, quantitative assessment of the effects of low-dose silica nanoparticles on lung cells: grasping complex toxicity with a great depth of field , 2015, BMC Genomics.

[28]  Wang,et al.  Replica Monte Carlo simulation of spin glasses. , 1986, Physical review letters.

[29]  R. Judson,et al.  The Toxicity Data Landscape for Environmental Chemicals , 2008, Environmental health perspectives.

[30]  Ivan Rusyn,et al.  Predicting drug-induced hepatotoxicity using QSAR and toxicogenomics approaches. , 2011, Chemical research in toxicology.

[31]  Michael W Deem,et al.  Parallel tempering: theory, applications, and new perspectives. , 2005, Physical chemistry chemical physics : PCCP.

[32]  Cinzia Nasuti,et al.  Dopaminergic system modulation, behavioral changes, and oxidative stress after neonatal administration of pyrethroids. , 2007, Toxicology.

[33]  Miao He,et al.  A Quantitative Toxicogenomics Assay for High-throughput and Mechanistic Genotoxicity Assessment and Screening of Environmental Pollutants. , 2016, Environmental science & technology.

[34]  T. Shioda,et al.  Prenatal Exposure to BPA Alters the Epigenome of the Rat Mammary Gland and Increases the Propensity to Neoplastic Development , 2014, PloS one.

[35]  K L Kolaja,et al.  Opportunities for Use of Human iPS Cells in Predictive Toxicology , 2011, Clinical pharmacology and therapeutics.

[36]  J. Castell,et al.  Mechanism-based selection of compounds for the development of innovative in vitro approaches to hepatotoxicity studies in the LIINTOP project. , 2010, Toxicology in vitro : an international journal published in association with BIBRA.

[37]  Tsuyoshi Kato,et al.  Kernel Classification Methods for Cancer Microarray Data , 2010 .

[38]  Craig Zwickl,et al.  An evaluation of in-house and off-the-shelf in silico models: implications on guidance for mutagenicity assessment. , 2015, Regulatory toxicology and pharmacology : RTP.

[39]  C. Léránth,et al.  Bisphenol A prevents the synaptogenic response to estradiol in hippocampus and prefrontal cortex of ovariectomized nonhuman primates , 2008, Proceedings of the National Academy of Sciences.

[40]  Wei Shi,et al.  BeadArray Expression Analysis Using Bioconductor , 2011, PLoS Comput. Biol..

[41]  Aixia Guo,et al.  Gene Selection for Cancer Classification using Support Vector Machines , 2014 .

[42]  D-S Cao,et al.  In silico toxicity prediction by support vector machine and SMILES representation-based string kernel , 2012, SAR and QSAR in environmental research.

[43]  Ayhan Demiriz,et al.  Semi-Supervised Support Vector Machines , 1998, NIPS.

[44]  S. Muresan,et al.  Chemical predictive modelling to improve compound quality , 2013, Nature Reviews Drug Discovery.

[45]  B. Weiss,et al.  Mercury exposure and child development outcomes. , 2004, Pediatrics.

[46]  Michal Linial,et al.  Using Bayesian Networks to Analyze Expression Data , 2000, J. Comput. Biol..

[47]  Ankur Omer,et al.  An overview of data mining algorithms in drug induced toxicity prediction. , 2014, Mini reviews in medicinal chemistry.

[48]  edited by Frank Emmert-Streib and Matthias Dehmer.  Medical Biostatistics for Complex Diseases , 1994 .

[49]  Xavier Robin,et al.  pROC: an open-source package for R and S+ to analyze and compare ROC curves , 2011, BMC Bioinformatics.

[50]  Andreas Bender,et al.  How Diverse Are Diversity Assessment Methods? A Comparative Analysis and Benchmarking of Molecular Descriptor Space , 2014, J. Chem. Inf. Model..

[51]  W. Fujibuchi,et al.  Effects of methylmercury exposure on neuronal differentiation of mouse and human embryonic stem cells. , 2012, Toxicology letters.

[52]  Senén Barro,et al.  Do we need hundreds of classifiers to solve real world classification problems? , 2014, J. Mach. Learn. Res..

[53]  Takehiro Hirai,et al.  Developing a practical toxicogenomics data analysis system utilizing open-source software. , 2013, Methods in molecular biology.

[54]  Toshihiko Ogura,et al.  Identification of a Primary Target of Thalidomide Teratogenicity , 2010, Science.

[55]  Norio Nakatsuji,et al.  Efficient establishment of human embryonic stem cell lines and long-term maintenance with stable karyotype by enzymatic bulk passage. , 2006, Biochemical and biophysical research communications.

[56]  Igor V. Tetko,et al.  Virtual Computational Chemistry Laboratory – Design and Description , 2005, J. Comput. Aided Mol. Des..

[57]  B. Ripley.  Support Functions and Datasets for Venables and Ripley's MASS , 2015 .

[58]  M. Jalali-Heravi,et al.  QSAR modelling of integrin antagonists using enhanced Bayesian regularised genetic neural networks , 2011, SAR and QSAR in environmental research.