{"paper_id":"3ef948eb-c291-4b61-a4d8-52c640e33677","body_text":"Chemical nature of attention deficit hyperactivity disorder (ADHD)- related chemical subfamily. | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Short Report Chemical nature of attention deficit hyperactivity disorder (ADHD)- related chemical subfamily. Masami Ishido This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-1701860/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract ・Using the bioassay based on rat hyperactivity, some endocrine disruptors were categorized into hyperactivity-associated group and hyperactivity-negative group. ・Two groups have differences in“Fraction sp3” (number of sp3-hybridized carbons/total carbon count) and the Tanimoto coefficient. ・A neural network model clearly classified the two groups. Random forest methods also showed the good prediction (R = 0.9, MAE (mean absolute error) = 0.06). ・Using a junction tree variational autoencoder, the core structure was interpolated between phthalate and phenol in the hyperactivity-associated group. Toxicology Environmental chemicals ADHD Toxicoinformatics Chemoinformatics Machine learning Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 1. Introduction Endocrine disruptors are defined as a diverse group of compounds that mimic the action of steroidal oestrogens. They include phenols and phthalate ester, such as bisphenol A and diethylhexyl phthalate (DEHP) and humans are exposed to these compounds through ingestion, inhalation, and dermal exposure because they are leaked from the food packaging (Ben-Jonathan and Steinmetz, 1998 ; Brotons, et al. 1995) or medical devices (NTP, 1998), respectively. Due to their estrogenic actions, the reproductive effects have been examined. Then, the public concern shifted to the possibility that endocrine disruptors might disrupt neuronal functions because there were a few reports that showed the transplacental transfer of many chemicals such as dioxins, bisphenol A and polychlorinated biphenyls (PCBs) from the mother to the fetus (Mori, 2001 ; Patandin, et al. 1994). Thus, the chemicals might exert their effects on the developing brain (Jacobson and Jacobson, 1996 ). Under these conditions, we hypothesized that endocrine disruptors might contribute to the incidence of neurodevelopmental disorders such as attention-deficit hyperactivity disorder (ADHD) and autism (Rice, 2000 ; London and Etzel, 2000 ). Hyperactivity occurs in the patients with ADHD and autism. A pioneering animal model of hyperactivity was produced by Shaywitz et al . (Shaywitz, et al. 1976 ). They showed that an intracisternal injection of 6-hydroxy dopamine (6-OHDA) to rat pups at 5 days of age resulted in increased motor activity between 2 and 4 weeks of age. Sixty-five chemicals are listed by the Ministry of the Environment, Government of Japan as possible endocrine disruptors that should be prioritized ( https://www.env.go.jp/ chemi/end/speed98/main/speed98-13.pdf, in Japanese). Using the protocol reported by Shaywitz et al. , we previously observed hyperactivity in the rats following the administration of endocrine disruptors, such as bisphenol A (Ishido, et al., 2004a ; Ishido, et al., 2007 ; Ishido, et al., 2011 ), p-octhylphenol (Masuo et al., 2004 ), nonylphenol (Masuo et al., 2004 ), dibutylphthalate (DBP)(Masuo et al., 2004 ), dicyclohexylphthalate (DCHP)(Ishido et al., 2004b ), diethylhexylphthalate (DEHP)(Ishido et al., 2005 ), and p-nitrotoluene (Ishido et al., 2004c ; 2007 ), concomitantly with impairment in immunoreactivity of tyrosine hydroxylase, a rate-limiting enzyme in catecholamine synthesis. Afterwards, epidemiological studies demonstrated the relationship between the levels of endocrine disruptors and the patients with ADHD (Kim et al., 2009 ; Cho 2010; Yolton et al., 2011 ; Harley et al. 2013 ; Chopra et al., 2014 ; Park et al., 2015 ; Huang et al., 2015 ; Tewar et al., 2016 ). Kim and colleagues (2009) reported a relationship between the urinary concentration of phthalic acid metabolites and ADHD (evaluated as a score by the teacher) in the school pupils in Korea. Notably, the study by Park and colleagues showed a relationship with the dopaminergic system in the brain ( Park et al., 2015 ). Tewar and colleagues (2016) investigated the relationship between the diagnostic criteria in DSM-IV (Diagnostic and Statistical Manual of Mental Disorders-V) and patients with ADHD with a caregiver and the urinary concentration of bisphenol A (Tewar et al., 2016 ). The half-life of bisphenol A is approximately 6 hours (and generally remains unchanged with multiple measurements), suggesting that the exposure is chronic and in equilibrium. We previously studied the effects of 18 endocrine disruptors on rat hyperactivity. Chemicals that elicited a significantly increased spontaneous motor activity compared to the control were defined as rat hyperactivity-associated chemicals, while those that elicited no increase were defined as negative chemicals. The use of chemoinformatics to reveal the chemical characteristics of both groups is rational since substantial progress has been achieved in the computational algorithm for examining the chemical aspects such as molecular descriptors and an open source chemoinformatics toolkit and a public database are fully available. Recently, machine learning has been applied to help propose new designs and synthetic routes because it is a very powerful tool for regression analyses and the classification and prediction of chemicals that I am evaluating. Here, I characterized the differences between hyperactivity-associated and hyperactivity-negative chemicals using cheminformatics and machine learning methods. 2. Methods 2.1. Screening of hyperactivity-associated chemicals in the rats. All animal experiments were performed in strict accordance with the Experiment and Related Activities in Academic Research Institutions guidelines, under the jurisdiction of the Ministry of Education, Culture, Sports, Science and Technology, Japan. The protocol was approved by the Committee on the Ethics of Animal Experiments of the National Institute for Environmental Studies (NIES), Japan. In addition, this study was conducted in compliance with the ARRIVE guidelines. Pregnant Wistar rats were obtained from Clea Japan (Tokyo, Japan). They were maintained in their home cages and provided a standard laboratory chow (MF diet, Oriental Yeast Corp., Tokyo, Japan) and distilled water ad libitum at 22 ℃ on a light-dark cycle (12 h/12 h) for at least one week. All animal care procedures were performed in accordance with NIES guidelines. Approximately 50 male pups were born to 10 pregnant rats and 5–7 pups were randomly housed. Male pups were selected to have a 10 ~ 14 g body weight at 5 days of age for screening tests. The rat pups were weaned at 3 weeks of age. Test chemicals were suspended in a minimal amount of 50% ethanol and brought to volume with olive oil. The chemical (87 nmol/10 µl/pup) was intracisternally administered to 5-day-old male pups (Ishido et al., 2004a ; 2004b ; 2004c ; 2005 ; 2011 ; Masuo et al., 2004 ). Control male rats were treated with vehicle alone (10 µl). At least, five pups were subjected to testing with one chemical or vehicle alone. The spontaneous motor activity of rats was individually measured in a home cage with a Supermex system (Muromachi Kikai, Tokyo, Japan) (Masuo et al., 1997 ; Ishido et al., 2002 ). In this system a sensor detects and measures the radiated body heat of an animal. A Supermex sensor head consists of paired infrared pyroelectric detectors. This system detects any object with a temperature at least 5 ℃ higher than the background within a cone-shaped area with a 6 m diameter and a 110°vertex. The sensor monitors motion in multiple zones of the cage through an array of Fresnel lenses placed above the cage and movement of the animal in the X-, Y-, and Z-axes can be determined. Activity was measured in 15 min increments for 22–24 h and animals were maintained on a 12 h light:dark cycle. Food and water were provided ad libitum at the beginning of counting, and rats were never disturbed in any way. Hyperactivity-associated chemicals were screened by measuring rat hyperactivity as an index, which was reflected by a significant increase in spontaneous motor activity in the rats. A statistically significant positive fold increase (p < 0.05) defined the hyperactivity-associated chemicals in this study, while nonhyperactivity-associated chemicals were defined as negative chemicals. Statistical analyses were conducted using repeated-measures analysis of variance (ANOVA) or Student’s t test with the StatView ver. 5.0 software (SAS Institute, Cary, NC) or R 4.2.0 software (public domain software). 2.2. Chemoinformatics Chemoinformatics was performed using RDKit, an open source chemoinformatics package. RDKit was installed from https://www.rdkit.org . Lipinski parameters for molecules were calculated using RDKit, including the molecular weight (MW), partition coefficient (MolLogP), the fraction of C atoms that are sp3 hybridized (FractionCSP3), and topological polar surface area (TPSA). The Morgan fingerprint, a variable-length fingerprint called Morgan2; ECFP4 (Morgan2) was used. The Tanimoto coefficient (Tc) was used to evaluate chemical similarity. It is calculated as the ratio between conserved features and the total number of features of each molecule. Tc is the fraction of features shared by A and B among the total number of features. The index range is from 0 to 1 . Tc = c/(a + b-c) = c /((a-c)+(b-c) + c) a: features of Compound A b: features of Compound B c: features common to Compounds A and B A dendrogram constructed using the Ward method was also provided by RDKit. 2.3. Machine learning The random forest and neural network models were created using molecular fingerprints as input. Predictions were established using RandomForestRegressor in the scikit-learn library. A neural network was constructed with Keras and TensorFlow ( Kanamaru 2018 ). A junction tree variational autoencoder (JT-VAE )(Kingma et al., 2014; Jin et al., 2019 ) was applied to interpolate two molecules, DPP and bisphenol A, using DGL (Deep Graph Library) and PyTorch, as described in https://github.com/shionhonda/dgl-playground/blob/master /jtnn.ipynb (Honda 2019 ). 3. Results 3.1. Molecular descriptors of hyperactivity-associated chemicals Eighteen chemicals were used to screen rat hyperactivity. Supplementary Fig. 1 shows the typical behavioral traits of the cases of dipentyl phthalate (DPP) and benzophenone. The hyperactivity-associated and the negative groups of chemicals are shown in Fig. 1 A. The hyperactivity-associated group included 11 types of chemicals, namely phenols (3 chemicals), phthalate esters (7 chemicals), and a nitro compound (1 chemical), while the negative group included benzophenone, possible metabolites of bisphenol A (2 chemicals), adipate, amitrol, and phthalate ester (1 chemical). The hyperactivity-associated group consisted of compounds containing 3 types of functional groups: phenol, ester and nitro functional groups (Fig. 1 B). Among the 11 hyperactivity-associated chemicals, DPP caused the highest levels of spontaneous motor activity in the rats. Then, molecular descriptors were analysed using RDKit. No noticeable differences were observed in the mean and variation of molecular weight (Fig. 2 A) between the two groups; therefore, comparisons were conducted for the other molecular indicators (Figs. 2 B to 2 D). A marked difference was observed in the fraction of hybridized electron orbitals (Fraction sp3; Fig. 2 C). In addition, the Tanimoto coefficient is used as an indicator to compare similarities among chemical substances. The Tanimoto coefficients of the two groups were compared (Fig. 3 ). DPP was used as the reference since it resulted in the highest spontaneous motor activity. Noticeable similarities were observed in the hyperactivity-associated group, with little similarity observed compared with the negative group. Moreover, the chemicals in the hyperactivity-associated group were dopaminergic neurotoxins that impair the development of the dopaminergic system in the brain(Ishido, et al., 2004a ; 2004b ; 2007 ;2017; Masuo et al., 2004 ;). Based on this result, an investigation was conducted to determine whether the substances were classified into the two known groups of potential dopaminergic neurotoxins (Table I)(Nagatsu 2002 ; Jayaraji 2016). Comparisons were performed using fingerprints similar to the Tanimoto index as an indicator. The group of dopaminergic neurotoxins containing amines was designated as DA amine, and the group of dopaminergic neurotoxins with chlorine was designated as DA oc. As shown in Fig. 4 a, the Tanimoto coefficient in the hyperactivity-associated group was much higher than that in the other groups. As shown in Fig. 4 B, these compounds were clearly categorized into three groups by principal component analysis (PCA). 3.2. Machine learning of hyperactivity-associated chemicals Machine learning was employed to further explore the chemical nature of hyperactivity-associated chemicals. A support vector machine failed to distinguish the hyperactivity-associated components from negative components based on the relationship between spontaneous motor activity and the Tanimoto coefficient (Supplementary Fig. 2). Neural networks succeeded in clearly distinguishing the two groups (Fig. 5 A) by gradually decreasing the loss function (Fig. 5 B); it was approximately zero at approximately 1,500 epochs (time steps; Fig. 5 B). Thus, machine learning was applicable to our dataset. Figure 6 shows the random forest prediction of spontaneous motor activity (SMA) based on the Tanimoto coefficient. A good correlation was observed between the observed SMA and predicted SMA (R = 0.9, mean absolute error (MAE) = 0.06). The activities of two chemicals were lower than basal rat spontaneous motor activity, showing a relatively low predictive accuracy. A junction tree variational autoencoder (JT-VAE) (Kingma et al. 2014; Jin et al., 2019 ) was employed between DPP and bisphenol A to deduce the presence of the core structure in the hyperactivity-associated group,. Figure 7 shows the possible interpolation of two types of chemicals, suggesting the presence of a shared core structure between the two chemicals. 4. Discussion There are several types of chemicals such as carcinogenic chemicals, nanomaterials, and epigenetic chemicals. In 1996, Dr. Theo Colborn et al . reported oestrogenic chemicals that disturb the endocrine system in wildlife and humans, called as endocrine disruptors (Colborn et al., 1996 ). Thus, reproductive organs should be investigated for endocrine disruptors. Due to the estrogenic activity of bisphenol A, its mode of action has been thought to be mediated by the estrogen receptor ER (Herz et al., 2017 ). At present, six target molecules with which bisphenol A can interact have been identified: ER alpha, estrogen-related receptor gamma (Takayanagi et al. 2017), pregnane X receptor (Sui et al. 2012 ), androgen receptor (Sohori et al., 1998), peroxisome proliferator activated receptor (Rui et al., 2011 ), and thyroid hormone receptor (Moriyama et al., 2002). Thus, this fact may explain why bisphenol A induces the dysfunction of many targets, leading to the recent concept of polypharmacology (Ellingson, et al., 2014 ). In addition to bisphenol A, the circumstances are similar for phthalates (Harris et al., 1997 ; Sun et al., 2018 ). In our screening of endocrine disruptors, bisphenol A and seven phthalates exerted the same effects on inducing nervous dysfunction, suggesting that unknown similarity in both chemical structures should exist. I employed the chemoinformatics and compared two groups to elucidate this property: hyperactivity-associated chemicals and hyperactivity-negative chemicals. First, I considered what would help distinguish between hyperactivity-associated and hyperactivity-negative compounds. Using a Ward dendrogram, individual compounds were unable to be clearly distinguished based on clustering distance (Supplementary Fig. 3). Therefore, I explored the differences in the physical properties of each group, such as molecular weight, fraction of hybridized electron orbitals (FractionCSP3), topological polar surface area, and number of rotatable bonds, and hydrogen bond donors and acceptors. Two distinctions -in FractionCSP3 and the Tanimoto coefficient were identified. Fraction CSP3 (Fsp3) indicates the complexity of the chemical, where Fsp3 = (number of sp3-hybridized carbons/total carbon count). Since the distribution of the molecular weight of both groups was not significantly different, the hyperactivity might be dependent on a higher number of sp3-hybridized carbons. The Tanimoto coefficient is an index of chemical similarity. The Tanimoto coefficient has two arguments: the query structure and the target structure. Dipentyl phthalate (DPP) was used as a reference since this compound resulted in the highest spontaneous motor activity (SMA) in our screen. The score of the hyperactivity-associated group was much higher than that of the negative group (Fig. 3 ). Although many dopaminergic toxins have been identified, as listed in Table I, hyperactivity-associated chemicals are distinguished in chemical similarity among the chemical family of dopaminergic toxins, as revealed by the Tanimoto coefficient (Fig. 4 ). Next, I considered why the hyperactivity-associated group contained three types of functional groups (Fig. 1 ). I first applied machine learning to classify those data (Fig. 5 ), establish a prediction (Fig. 6 ), and explore the potential hidden backbone of the chemical structure; the results indicated that machine learning was applicable to those datasets. We then used a junction tree variational autoencoder (JT-VAE) (Kingma et al., 2014; Jin et al., 2019 ). JT-VAE is a generative model that utilizes deep neural networks to describe the distribution of observed and latent (unobserved) variables. Using JT-VAE, the latent space was interpolated between DPP and bisphenol A (Fig. 7 ). Continuous interpolation seems to be performed, suggesting that a common backbone of these molecules might contribute to hyperactivity in the rats. Hopefully, the application of the chemical descriptors obtained in this study will facilitate chemical development to replace the hyperactivity-associated chemicals and will help elucidate the etiology of ADHD in the human. 5. Conclusion In this study, we screened some endocrine disruptors based on rat hyperactivity, resulting in two groups: hyperactivity-associated group and hyperactivity-negative group. There were differences in“Fraction sp3” (number of sp3-hybridized carbons/total carbon count) and the Tanimoto coefficient between two groups. Our findings will help to develop more safer chemical and to elucidate the etiology of ADHD in the human. Declarations Acknowledgements The author thanks Drs. Ayako Furuhama, Takashi Kanamaru, Shion Honda, and Bibhash Chandra Mitra for consultations regarding chemoinformatics. This work was supported by a Grant-in-Aid for Scientific Research (KAKENHI). Competing interests The author has no conflicts of interest to declare. References Ben-Jonathan, N., Steinmetz, R., 1998. Xenoestrogens: the emerging story of bisphenol A. Trend Endocrinol. Metab. 9:124-128. Brotons, J.A., et al. Xenoestrogens released from lacquer coatings in food cans. Environ. Health Perspect. 103:608-612 (1995). Cho, S-C. et al. 2010. Relationship between environmental phthalate exposure and the intelligence of school-age children. Environ. Health Perspect. 118, 1027-1032. Chopra, V., Harley, K.,Lahiff, M., Eskenazia, B., 2014. Association between phthalates and attention deficit disorder and learning disability in U.S. children, 6–15 years. Environ Res. 128: 64–69. Colborn, T., Dumanoski, D., Meyers, J.P., 1996. Our stolen future. (Dutton, New York) Ellingson, S.R., Smith, J.C., Baudry, J., 2014. Polypharmacology and supercomputer-based docking: opportunities and challenges. Mol. Simulation 40, 848-854. Harris, C.A., Henttu, P., Parker, M.G., Sumpter, J.P., 1997. The estrogenic activity of phthalate esters in vitro . Environ. Health Perspect. 105, 802-811. Harley, K.G. et al. , 2013. Prenatal and postnatal bisphenol A and body mass index in childhood in the CHAMACOS cohort. Environ Health Perspect , 121(4): 514-520. Herz, C., et al., 2017. Low-dose levels of bisphenol A inhibit telomerase via ER/GPR30-ERK signalling, impair DNA integrity and reduce cell proliferation in primary PBMC. Sci. Reports 7, doi:10.1038/s41598-017-15978-2. Honda, S., 2019 Interpolation with pre-trained JTNN . https:// github.com/ shionhonda/ dgl-playground/blob/master/jtnn.ipynb. (accessed 6 April 2022) Huang, H-B. et al. , 2015.Fetal and Childhood Exposure to Phthalate Diesters and Cognitive Function in Children Up to 12 Years of Age: Taiwanese Maternal and Infant Cohort Study. PLOS One https://doi.org/10.1371/journal.pone.0131910 Ishido, M., Masuo, Y., Oka, S., Kunimoto, M., Morita, M., 2002. Application of Supermex system to screen behavioral traits produced by tributyltin in the rat. J. Health Sci. 48, 451-454. Ishido, M., Masuo, Y., Kunimoto, M., Oka, S., Morita, M., 2004a. Bisphenol A causes hyperactivity in the rat concomitantly with impairment of tyrosine hydroxylase immunoreactivity. J. Neurosci. Res. 76, 423-433. Ishido, M., Masuo, Y., Oka S, Niki E, Morita, M. 2004b. p -Nitrotoluene causes hyperactivity in the rat. Neurosci. Lett . 366, 1-5. Ishido, M., et al., 2004c. Dicyclohexylphthalate causes hyperactivity in the rat concomitantly with impairment of tyrosine hydroxylase immunoreactivity, J. Neurochem. 91, 69-76. Ishido, M., Morita, M., Oka, S., Masuo Y., 2005.Alteration of gene expression of G-protein-coupled receptors in endocrine disruptors-caused hyperactive rats. Regulatory Peptides 126, 145-153. Ishido, M., Yonemoto, J., Morita, M., 2007. Mesencephalic neurodegeneration in the orally administered bisphenol A-caused hyperactive rats. Toxicol. Lett . 173, 66-72. Ishido, M., Masuo, Y., Terasaki, M., Morita, M., 2011. Rat hyperactivity by bisphenol A, but not by its derivatives, 3-hydroxybisphenol A or bisphenol A 3,4-quinone. Toxicol. Lett. 206, 300-305. Ishido, M., Usu, R., 2017. Orally administered p -nitrotoluene causes hyperactivity, concomitantly with gliosis and impairment of tyrosine hydroxylase immunoreactivity in the rat substantia nigra. Fund. Toxicol. Sci. 4, 151-158. Jacobson, J.L., Jacobson, S.W., 1996. Intellectual impairment in children exposed to polychlorinated biphenyls in utero. New Engl. J. Med. 335:783-789. Jayaraj, R., Megha, P., Sreedev, P., 2016. Organochlorine pesticides, their toxic effects on living organisms and their fate in the environment. Interdiscip. Toxicol. 9, 90-100. Jin, W., Barzilay, R., Jaakkola, T., 2019. Junctional tree variational autoencoder for molecular graph generation. arXiv Prepr. arXiv: 1802.04364v4 . Kanamaru, T., 2018. Machine learning with Raspberry Pi. (Kodansha, Tokyo) ( in Japanese ) Kingma, D., Welling, M., 2014. Auto-encoding variational bays. arXiv Prepr. arXiv1312.6114. Kim, B-N. et al. , 2009. Phthalates exposure and attention-deficit/hyperactivity disorder in school-age children. Biol Psychiatry 66, 958-63. London, F., Etzel, R.A., 2000. The environment as an etiologic factor in autism: a new direction for research. Environ. Health Persp. 108 (Suppl 3):401-404. Masuo Y, Matsumoto Y, Morita S, Noguchi J., 1997. A novel method for counting spontaneous motor activity in the rat. Brain Res, Protocols 1, 321-326. Masuo, Y., Ishido, M., Morita, M., Oka, S., 2004. Effects of neonatal treatment with 6-hydroxydopamine and endocrine disruptors on motor activity and gene expression in rats. Neural Plasticity 11, 59-76. Miodovnik, A. et al., 2011. Endocrine disruptors and childhood social impairment. Neurotoxicology. 32, 261–267. Mori, C., 2001. Assessment of fetal environmental exposure and subtle male reproductive disorders in Japan. In Proceedings of the 1st U.S.-Japan workshop on endocrine disrupting chemicals and their toxicological evaluation. 103-106. Moriyama, K. et. al. , 2002. Thyroid hormone action is disrupted by bisphenol A as an antagonist. J. Clin. Endocrinol. Metab. 87, 5185-5190. Nagatsu, T., 2002. Amine-related neurotoxins in Parkinson’s disease: Past, present, and future. Neurotoxicol. Teratology 24, 565-569. NTP (National Toxicology Program), 1998. Report on carcinogens , eighth edition. National Toxicology Program: National Institutes of Health (NIH) publications, Research Triangle, Park, NC, USA. Park, S. et al. , 2015. Association between phthalates and externalizing behaviors and cortical thickness in children with attention deficit hyperactivity disorder. Psychol Med 45, 1601-12. doi: 10.1017/S0033291714002694. Patandin, S. et al., 1994. Effects of environmental exposure to polychlorinated biphenyls and dioxins on cognitive abilities in Dutch children at 42 months of age. J. Pediatr 134 : 33-41. Polanska, K., Ligocka, D., Sobala, W., Hanke, W., 2014. Phthalate exposure and child development: the Polish Mother and Child Cohort Study. Early Hum. Dev. 90, 477-485. Rice, D.C., 2000. Parallels between attention deficit hyperactivity disorder and behavioral deficits produced by neurotoxic exposure in monkeys. Environ. Health Persp. 108(Suppl 3):405-408. Shaywitz, B.A., Yager, R.D., Klopper, J.H., 1976. Selective brain dopamine depletion in developing rats: an experimental model of minimal brain dysfunction. Science 191, 305-308. Sohoni, P., Sumpter, J.P., 1998. Several environmental oestrogens are also anti-androgens. J. Endocrinol. 158, 327-339. Stein, P., et al. , 2015. Bisphenol A exposure in children with autism spectrum disorders. Autism Res. 8, 272-83. doi: 10.1002/aur.1444. Sui, Y. et al. , 2012. Bisphenol A and its analogues activate human pregnane X receptor. Environ. Health Perspect. 120, 399-405. Sun, D. et al. , 2018. Effect of di-(2-ethylhexyl)phthalate on the hypothalamus-pituitary-thyroid axis in adolescent rat. Endocrine J. 65, 261-268. Takayanagi, S. et al. , 2006. Endocrine disruptor bisphenol A strongly binds to human estrogen-related receptor gamma (ERR gamma) with high constitutive activity. Toxicol. Lett. 167, 95-105. Tewar, S. et al. , 2016. Association of Bisphenol A exposure and Attention-Deficit/Hyperactivity Disorder in a national sample of U.S. children. Environ. Res. 150:112-118. doi: 10.1016/j.envres.2016.05.040. Rui, A. et al., 2011. Peroxisome proliferator-activated receptor gamma is a target for halogenated analogs of bisphenol A. Environ. Health Perspect. 119, 1227-1232. Yolton, K. et al. , 2011. Prenatal exposure to bisphenol A and phthalates and infant neurobehavior. Neurotoxicol. Teratol. 33, 558-66. Table Table I . Potential dopaminergic toxins Chemicals Amine MPP+ TaClo-d4 Norharman(βCarboline) 1,2,3,4-Tetrahydroisoquinoline 1-Phenyl-1,2,3,4-tetrahydroisoquinoline Salsolinol（1-Methyl-6,7-dihydroxytetrahydroisoquinoline） Norsalsolinol Organochlorines DDT Lindane Dieldrin Heptachlor Methoxychlor Trichloroethylene Aldrin Chlordane Endosulfan Supplementary Files 2022.4.28M.Ishidosupplalone.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {\"props\":{\"pageProps\":{\"initialData\":{\"identity\":\"rs-1701860\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":true,\"archivedVersions\":[],\"articleType\":\"Short Report\",\"associatedPublications\":[],\"authors\":[{\"id\":109410503,\"identity\":\"5e9f643d-d4ab-4d73-a01e-cd894e27f59f\",\"order_by\":0,\"name\":\"Masami Ishido\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA0ElEQVRIie3QMQrCMBTG8S8U6hLtmkmvkCJ0sYdpl3apzh0FoZPoKughuolb5a0VD+HaIeAi0sEqOjil3UTyJ1N4P8ILYDL9YDY4Cg4MYb1v2LwlGbcnaMjzhPrBTwPndCyqlOJ9zyKFuw9rq9G2mAXHXUnTw8KOBFtFYLtCR7ikfkbTnLgHtiSwTaAhTvkisSTnqtoRJC8SSOIQuLUhIpHNLrGbk+2JcB5x7S6jdTlWVToZyTNdlKr9oav7se/CDNzddBFA3bwruhGTyWT6/x6ojkMN1GisfgAAAABJRU5ErkJggg==\",\"orcid\":\"\",\"institution\":\"National Institute for Environmental Studies\",\"correspondingAuthor\":true,\"submittingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Masami\",\"middleName\":\"\",\"lastName\":\"Ishido\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2022-05-28 06:38:28\",\"currentVersionCode\":1,\"declarations\":{\"humanSubjects\":false,\"vertebrateSubjects\":true,\"conflictsOfInterestStatement\":true,\"humanSubjectEthicalGuidelines\":false,\"humanSubjectConsent\":false,\"humanSubjectClinicalTrial\":false,\"humanSubjectCaseReport\":false,\"vertebrateSubjectEthicalGuidelines\":true,\"coiExplicitlySet\":false},\"doi\":\"10.21203/rs.3.rs-1701860/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-1701860/v1\",\"draftVersion\":[],\"editorialEvents\":[],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":22093717,\"identity\":\"f52857a3-e7d4-4abe-b277-fe7aae3bace1\",\"added_by\":\"auto\",\"created_at\":\"2022-05-31 18:28:18\",\"extension\":\"jpg\",\"order_by\":1,\"title\":\"Figure 1\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":44077,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003e\\u003cem\\u003eA; \\u003c/em\\u003eThe chemical structures of hyperactivity-associated (\\u003c/strong\\u003e\\u003cem\\u003eleft panel\\u003c/em\\u003e\\u003cstrong\\u003e) and negative (\\u003c/strong\\u003e\\u003cem\\u003eright panel\\u003c/em\\u003e\\u003cstrong\\u003e) chemicals in the rat hyperactivity screening. \\u003c/strong\\u003eA screen of the ability of the chemicals to induce hyperactivity in rats was conducted as described in the Methods.\\u003c/p\\u003e\\u003cp\\u003e\\u003cstrong\\u003e\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\u003cp\\u003e\\u003cstrong\\u003e\\u003cem\\u003eB\\u003c/em\\u003e; Three types of functional groups in hyperactivity-associated chemicals. \\u003c/strong\\u003eGasteiger partial charges of typical hyperactivity-associated chemicals were determined using RDKit, as indicated.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"f1.jpg\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-1701860/v1/0e7cd94a41ca68cb377d44dc.jpg\"},{\"id\":22093917,\"identity\":\"7ff4bbe0-0992-4750-8c82-c2404c77d167\",\"added_by\":\"auto\",\"created_at\":\"2022-05-31 18:33:18\",\"extension\":\"jpg\",\"order_by\":2,\"title\":\"Figure 2\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":46248,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eComparison of chemical descriptors between hyperactivity-associated chemicals and negative chemicals. \\u003c/strong\\u003eEighteen chemicals were identified in the rat hyperactivity screen, including 11 hyperactivity-associated chemicals and 7 negative chemicals. The chemical descriptors of the two chemical groups were analysed using RDKit, an open source chemoinformatics toolkit. (\\u003cem\\u003ea\\u003c/em\\u003e, Distribution of molecular weight; \\u003cem\\u003eb\\u003c/em\\u003e, hydrophobicity (MolLogP); \\u003cem\\u003ec\\u003c/em\\u003e, fraction of sp3-hybridized electron orbitals; and \\u003cem\\u003ed\\u003c/em\\u003e, topological polar surface area (TPSA)). ＊; p=0.022.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"f2.jpg\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-1701860/v1/b706681ec55b27bd0910430d.jpg\"},{\"id\":22093715,\"identity\":\"82a4f18b-0eac-44ba-8dfc-32b135e0e238\",\"added_by\":\"auto\",\"created_at\":\"2022-05-31 18:28:18\",\"extension\":\"jpg\",\"order_by\":3,\"title\":\"Figure 3\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":28069,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eComparison of chemical similarity between the two groups based on the Tanimoto coefficient. \\u003c/strong\\u003eThe\\u003cstrong\\u003e \\u003c/strong\\u003eTanimoto coefficient was calculated based on the phthalate ester DPP (the hyperactivity-associated chemical causing the highest spontaneous motor activity), as presented in a box plot. ＊; p=0.007.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"f3.jpg\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-1701860/v1/5f739becbfcbd11e73c15bb4.jpg\"},{\"id\":22093719,\"identity\":\"949ad236-528b-4be4-9230-34c100a442b7\",\"added_by\":\"auto\",\"created_at\":\"2022-05-31 18:28:18\",\"extension\":\"jpg\",\"order_by\":4,\"title\":\"Figure 4\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":40696,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003ePrincipal component analysis (PCA) for classifying the three toxic groups of chemicals towards dopaminergic neurons. \\u003c/strong\\u003ePCA was performed among the three chemical groups based on the Tanimoto coefficient, as indicated. \\u003cem\\u003eDA amine\\u003c/em\\u003e; dopaminergic toxins that contain amines, \\u003cem\\u003eDA oc\\u003c/em\\u003e; dopaminergic toxins that contain chlorine. ＊; p=0.007.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"f4.jpg\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-1701860/v1/fb6bce237979b7677d380f80.jpg\"},{\"id\":22093721,\"identity\":\"ad78ebe3-8ce6-4f2c-886e-b16441ab73cc\",\"added_by\":\"auto\",\"created_at\":\"2022-05-31 18:28:18\",\"extension\":\"jpg\",\"order_by\":5,\"title\":\"Figure 5\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":33257,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eClassification of the two chemical groups using neural networks. \\u003c/strong\\u003eThe Tanimoto coefficient was calculated based on the phthalate ester DPP (the hyperactivity-associated chemical causing the highest spontaneous motor activity), and the results are presented in a scatter plot versus spontaneous motor activity (SMA) with the classification by neural networks shown in different coloured backgrounds (\\u003cem\\u003ea\\u003c/em\\u003e). Note that negative compounds (\\u003cem\\u003epurple\\u003c/em\\u003e) and hyperactivity-associated compounds (\\u003cem\\u003ered\\u003c/em\\u003e) are shown with light blue and orange backgrounds, respectively. The loss function is shown in (\\u003cem\\u003eb\\u003c/em\\u003e).\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"f5.jpg\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-1701860/v1/0759413e2c36e2e1f1a6e6de.jpg\"},{\"id\":22093718,\"identity\":\"f99c4214-b689-455b-aadf-075afc3461e1\",\"added_by\":\"auto\",\"created_at\":\"2022-05-31 18:28:18\",\"extension\":\"jpg\",\"order_by\":6,\"title\":\"Figure 6\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":20177,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003ePrediction of spontaneous motor activity based on the Tanimoto coefficient using a random forest analysis. \\u003c/strong\\u003eDatasets for chemicals screened for their ability to induce hyperactivity in rats were subjected to a random forest analysis. Notably, the activities of two chemicals were lower than basal rat spontaneous motor activity, showing a relatively lower predictive accuracy.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"f6.jpg\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-1701860/v1/86653e285bcf65fafd7e1e15.jpg\"},{\"id\":22093722,\"identity\":\"56bf2273-bdb8-4f6f-843a-ee0ee83de068\",\"added_by\":\"auto\",\"created_at\":\"2022-05-31 18:28:18\",\"extension\":\"jpg\",\"order_by\":7,\"title\":\"Figure 7\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":54022,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eInterpolation of between DPP and bisphenol A by a junction tree variational autoencoder (JT-VAE). \\u003c/strong\\u003e\\u003c/p\\u003e\\u003cp\\u003e\\u003cstrong\\u003e\\u0026nbsp;\\u0026nbsp;\\u0026nbsp;\\u0026nbsp;\\u0026nbsp;\\u003c/strong\\u003eSMILES structures for DPP and bisphenol A were subjected to the encoder, followed by interpolations in latent space. Then, they were decoded, and 50 possible chemicals are shown.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"f7.jpg\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-1701860/v1/53add9bbfb1fcd2cea6c09a8.jpg\"},{\"id\":22093919,\"identity\":\"890ba412-b194-476c-8af9-6ccaab5184fa\",\"added_by\":\"auto\",\"created_at\":\"2022-05-31 18:33:20\",\"extension\":\"pdf\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":643902,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-1701860/v1/829e60b3-9a9e-4189-84e5-d62df7176f13.pdf\"},{\"id\":22093918,\"identity\":\"0bc3879d-b56c-4c81-94c3-dc1f81495e4c\",\"added_by\":\"auto\",\"created_at\":\"2022-05-31 18:33:18\",\"extension\":\"docx\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":240392,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"2022.4.28M.Ishidosupplalone.docx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-1701860/v1/f7228a6f406dd69c0851751c.docx\"}],\"financialInterests\":\"\",\"formattedTitle\":\"\\u003cp\\u003eChemical nature of attention deficit hyperactivity disorder (ADHD)- related chemical subfamily. \\u003c/p\\u003e\",\"fulltext\":[{\"header\":\"1. Introduction\",\"content\":\"\\u003cp\\u003eEndocrine disruptors are defined as a diverse group of compounds that mimic the action of steroidal oestrogens. They include phenols and phthalate ester, such as bisphenol A and diethylhexyl phthalate (DEHP) and humans are exposed to these compounds through ingestion, inhalation, and dermal exposure because they are leaked from the food packaging (Ben-Jonathan and Steinmetz, \\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1998\\u003c/span\\u003e; Brotons, et al. 1995) or medical devices (NTP, 1998), respectively. Due to their estrogenic actions, the reproductive effects have been examined.\\u003c/p\\u003e \\u003cp\\u003eThen, the public concern shifted to the possibility that endocrine disruptors might disrupt neuronal functions because there were a few reports that showed the transplacental transfer of many chemicals such as dioxins, bisphenol A and polychlorinated biphenyls (PCBs) from the mother to the fetus (Mori, \\u003cspan citationid=\\\"CR30\\\" class=\\\"CitationRef\\\"\\u003e2001\\u003c/span\\u003e; Patandin, et al. 1994). Thus, the chemicals might exert their effects on the developing brain (Jacobson and Jacobson, \\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e1996\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eUnder these conditions, we hypothesized that endocrine disruptors might contribute to the incidence of neurodevelopmental disorders such as attention-deficit hyperactivity disorder (ADHD) and autism (Rice, \\u003cspan citationid=\\\"CR37\\\" class=\\\"CitationRef\\\"\\u003e2000\\u003c/span\\u003e; London and Etzel, \\u003cspan citationid=\\\"CR26\\\" class=\\\"CitationRef\\\"\\u003e2000\\u003c/span\\u003e). Hyperactivity occurs in the patients with ADHD and autism. A pioneering animal model of hyperactivity was produced by Shaywitz \\u003cem\\u003eet al\\u003c/em\\u003e. (Shaywitz, et al. \\u003cspan citationid=\\\"CR38\\\" class=\\\"CitationRef\\\"\\u003e1976\\u003c/span\\u003e). They showed that an intracisternal injection of 6-hydroxy dopamine (6-OHDA) to rat pups at 5 days of age resulted in increased motor activity between 2 and 4 weeks of age.\\u003c/p\\u003e \\u003cp\\u003eSixty-five chemicals are listed by the Ministry of the Environment, Government of Japan as possible endocrine disruptors that should be prioritized (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://www.env.go.jp/\\u003c/span\\u003e\\u003cspan address=\\\"https://www.env.go.jp/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e chemi/end/speed98/main/speed98-13.pdf, in Japanese). Using the protocol reported by Shaywitz \\u003cem\\u003eet al.\\u003c/em\\u003e, we previously observed hyperactivity in the rats following the administration of endocrine disruptors, such as bisphenol A (Ishido, et al., \\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e2004a\\u003c/span\\u003e; Ishido, et al., \\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e2007\\u003c/span\\u003e; Ishido, et al., \\u003cspan citationid=\\\"CR18\\\" class=\\\"CitationRef\\\"\\u003e2011\\u003c/span\\u003e), p-octhylphenol (Masuo et al., \\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e2004\\u003c/span\\u003e), nonylphenol (Masuo et al., \\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e2004\\u003c/span\\u003e), dibutylphthalate (DBP)(Masuo et al., \\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e2004\\u003c/span\\u003e), dicyclohexylphthalate (DCHP)(Ishido et al., \\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e2004b\\u003c/span\\u003e), diethylhexylphthalate (DEHP)(Ishido et al., \\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e2005\\u003c/span\\u003e), and p-nitrotoluene (Ishido et al., \\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e2004c\\u003c/span\\u003e; \\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e2007\\u003c/span\\u003e), concomitantly with impairment in immunoreactivity of tyrosine hydroxylase, a rate-limiting enzyme in catecholamine synthesis.\\u003c/p\\u003e \\u003cp\\u003eAfterwards, epidemiological studies demonstrated the relationship between the levels of endocrine disruptors and the patients with ADHD (Kim et al., \\u003cspan citationid=\\\"CR25\\\" class=\\\"CitationRef\\\"\\u003e2009\\u003c/span\\u003e; Cho 2010; Yolton et al., \\u003cspan citationid=\\\"CR46\\\" class=\\\"CitationRef\\\"\\u003e2011\\u003c/span\\u003e; Harley et al. \\u003cspan citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e2013\\u003c/span\\u003e; Chopra et al., \\u003cspan citationid=\\\"CR4\\\" class=\\\"CitationRef\\\"\\u003e2014\\u003c/span\\u003e; Park et al., \\u003cspan citationid=\\\"CR34\\\" class=\\\"CitationRef\\\"\\u003e2015\\u003c/span\\u003e; Huang et al., \\u003cspan citationid=\\\"CR11\\\" class=\\\"CitationRef\\\"\\u003e2015\\u003c/span\\u003e; Tewar et al., \\u003cspan citationid=\\\"CR44\\\" class=\\\"CitationRef\\\"\\u003e2016\\u003c/span\\u003e). Kim and colleagues (2009) reported a relationship between the urinary concentration of phthalic acid metabolites and ADHD (evaluated as a score by the teacher) in the school pupils in Korea. Notably, the study by Park and colleagues showed a relationship with the dopaminergic system in the brain ( Park et al., \\u003cspan citationid=\\\"CR34\\\" class=\\\"CitationRef\\\"\\u003e2015\\u003c/span\\u003e). Tewar and colleagues (2016) investigated the relationship between the diagnostic criteria in DSM-IV (Diagnostic and Statistical Manual of Mental Disorders-V) and patients with ADHD with a caregiver and the urinary concentration of bisphenol A (Tewar et al., \\u003cspan citationid=\\\"CR44\\\" class=\\\"CitationRef\\\"\\u003e2016\\u003c/span\\u003e). The half-life of bisphenol A is approximately 6 hours (and generally remains unchanged with multiple measurements), suggesting that the exposure is chronic and in equilibrium.\\u003c/p\\u003e \\u003cp\\u003eWe previously studied the effects of 18 endocrine disruptors on rat hyperactivity. Chemicals that elicited a significantly increased spontaneous motor activity compared to the control were defined as rat hyperactivity-associated chemicals, while those that elicited no increase were defined as negative chemicals.\\u003c/p\\u003e \\u003cp\\u003eThe use of chemoinformatics to reveal the chemical characteristics of both groups is rational since substantial progress has been achieved in the computational algorithm for examining the chemical aspects such as molecular descriptors and an open source chemoinformatics toolkit and a public database are fully available.\\u003c/p\\u003e \\u003cp\\u003eRecently, machine learning has been applied to help propose new designs and synthetic routes because it is a very powerful tool for regression analyses and the classification and prediction of chemicals that I am evaluating.\\u003c/p\\u003e \\u003cp\\u003eHere, I characterized the differences between hyperactivity-associated and hyperactivity-negative chemicals using cheminformatics and machine learning methods.\\u003c/p\\u003e\"},{\"header\":\"2. Methods\",\"content\":\"\\u003cdiv id=\\\"Sec3\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.1. Screening of hyperactivity-associated chemicals in the rats.\\u003c/h2\\u003e \\u003cp\\u003e All animal experiments were performed in strict accordance with the Experiment and Related Activities in Academic Research Institutions guidelines, under the jurisdiction of the Ministry of Education, Culture, Sports, Science and Technology, Japan. The protocol was approved by the Committee on the Ethics of Animal Experiments of the National Institute for Environmental Studies (NIES), Japan. In addition, this study was conducted in compliance with the ARRIVE guidelines. Pregnant Wistar rats were obtained from Clea Japan (Tokyo, Japan). They were maintained in their home cages and provided a standard laboratory chow (MF diet, Oriental Yeast Corp., Tokyo, Japan) and distilled water ad libitum at 22 ℃ on a light-dark cycle (12 h/12 h) for at least one week. All animal care procedures were performed in accordance with NIES guidelines. Approximately 50 male pups were born to 10 pregnant rats and 5\\u0026ndash;7 pups were randomly housed. Male pups were selected to have a 10\\u0026thinsp;~\\u0026thinsp;14 g body weight at 5 days of age for screening tests. The rat pups were weaned at 3 weeks of age.\\u003c/p\\u003e \\u003cp\\u003eTest chemicals were suspended in a minimal amount of 50% ethanol and brought to volume with olive oil. The chemical (87 nmol/10 \\u0026micro;l/pup) was intracisternally administered to 5-day-old male pups (Ishido et al., \\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e2004a\\u003c/span\\u003e; \\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e2004b\\u003c/span\\u003e; \\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e2004c\\u003c/span\\u003e;\\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e2005\\u003c/span\\u003e; \\u003cspan citationid=\\\"CR18\\\" class=\\\"CitationRef\\\"\\u003e2011\\u003c/span\\u003e; Masuo et al., \\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e2004\\u003c/span\\u003e). Control male rats were treated with vehicle alone (10 \\u0026micro;l). At least, five pups were subjected to testing with one chemical or vehicle alone.\\u003c/p\\u003e \\u003cp\\u003eThe spontaneous motor activity of rats was individually measured in a home cage with a Supermex system (Muromachi Kikai, Tokyo, Japan) (Masuo et al., \\u003cspan citationid=\\\"CR27\\\" class=\\\"CitationRef\\\"\\u003e1997\\u003c/span\\u003e; Ishido et al., \\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e2002\\u003c/span\\u003e). In this system a sensor detects and measures the radiated body heat of an animal. A Supermex sensor head consists of paired infrared pyroelectric detectors. This system detects any object with a temperature at least 5 ℃ higher than the background within a cone-shaped area with a 6 m diameter and a 110\\u0026deg;vertex. The sensor monitors motion in multiple zones of the cage through an array of Fresnel lenses placed above the cage and movement of the animal in the X-, Y-, and Z-axes can be determined. Activity was measured in 15 min increments for 22\\u0026ndash;24 h and animals were maintained on a 12 h light:dark cycle. Food and water were provided ad libitum at the beginning of counting, and rats were never disturbed in any way.\\u003c/p\\u003e \\u003cp\\u003eHyperactivity-associated chemicals were screened by measuring rat hyperactivity as an index, which was reflected by a significant increase in spontaneous motor activity in the rats. A statistically significant positive fold increase (p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05) defined the hyperactivity-associated chemicals in this study, while nonhyperactivity-associated chemicals were defined as negative chemicals. Statistical analyses were conducted using repeated-measures analysis of variance (ANOVA) or Student\\u0026rsquo;s t test with the \\u003cem\\u003eStatView\\u003c/em\\u003e ver. 5.0 software (SAS Institute, Cary, NC) or R 4.2.0 software (public domain software).\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec4\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.2. Chemoinformatics\\u003c/h2\\u003e \\u003cp\\u003eChemoinformatics was performed using RDKit, an open source chemoinformatics package. RDKit was installed from \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://www.rdkit.org\\u003c/span\\u003e\\u003cspan address=\\\"https://www.rdkit.org\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e. Lipinski parameters for molecules were calculated using RDKit, including the molecular weight (MW), partition coefficient (MolLogP), the fraction of C atoms that are sp3 hybridized (FractionCSP3), and topological polar surface area (TPSA).\\u003c/p\\u003e \\u003cp\\u003eThe Morgan fingerprint, a variable-length fingerprint called Morgan2; ECFP4 (Morgan2) was used. The Tanimoto coefficient (Tc) was used to evaluate chemical similarity. It is calculated as the ratio between conserved features and the total number of features of each molecule. Tc is the fraction of features shared by A and B among the total number of features. The index range is from 0 to 1 .\\u003c/p\\u003e \\u003cp\\u003eTc\\u0026thinsp;=\\u0026thinsp;c/(a\\u0026thinsp;+\\u0026thinsp;b-c)\\u0026thinsp;=\\u0026thinsp;c /((a-c)+(b-c)\\u0026thinsp;+\\u0026thinsp;c)\\u003c/p\\u003e \\u003cp\\u003ea: features of Compound A\\u003c/p\\u003e \\u003cp\\u003eb: features of Compound B\\u003c/p\\u003e \\u003cp\\u003ec: features common to Compounds A and B\\u003c/p\\u003e \\u003cp\\u003eA dendrogram constructed using the Ward method was also provided by RDKit.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec5\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.3. Machine learning\\u003c/h2\\u003e \\u003cp\\u003eThe random forest and neural network models were created using molecular fingerprints as input. Predictions were established using RandomForestRegressor in the scikit-learn library. A neural network was constructed with Keras and TensorFlow ( Kanamaru \\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e2018\\u003c/span\\u003e). A junction tree variational autoencoder (JT-VAE )(Kingma et al., 2014; Jin et al., \\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e2019\\u003c/span\\u003e) was applied to interpolate two molecules, DPP and bisphenol A, using DGL (Deep Graph Library) and PyTorch, as described in \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://github.com/shionhonda/dgl-playground/blob/master\\u003c/span\\u003e\\u003cspan address=\\\"https://github.com/shionhonda/dgl-playground/blob/master\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e /jtnn.ipynb (Honda \\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e2019\\u003c/span\\u003e).\\u003c/p\\u003e \\u003c/div\\u003e\"},{\"header\":\"3. Results\",\"content\":\"\\u003cdiv id=\\\"Sec7\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.1. Molecular descriptors of hyperactivity-associated chemicals\\u003c/h2\\u003e \\u003cp\\u003eEighteen chemicals were used to screen rat hyperactivity. Supplementary Fig.\\u0026nbsp;1 shows the typical behavioral traits of the cases of dipentyl phthalate (DPP) and benzophenone. The hyperactivity-associated and the negative groups of chemicals are shown in Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eA. The hyperactivity-associated group included 11 types of chemicals, namely phenols (3 chemicals), phthalate esters (7 chemicals), and a nitro compound (1 chemical), while the negative group included benzophenone, possible metabolites of bisphenol A (2 chemicals), adipate, amitrol, and phthalate ester (1 chemical).\\u003c/p\\u003e \\u003cp\\u003eThe hyperactivity-associated group consisted of compounds containing 3 types of functional groups: phenol, ester and nitro functional groups (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eB). Among the 11 hyperactivity-associated chemicals, DPP caused the highest levels of spontaneous motor activity in the rats.\\u003c/p\\u003e \\u003cp\\u003eThen, molecular descriptors were analysed using RDKit. No noticeable differences were observed in the mean and variation of molecular weight (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eA) between the two groups; therefore, comparisons were conducted for the other molecular indicators (Figs.\\u0026nbsp;\\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eB to \\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eD). A marked difference was observed in the fraction of hybridized electron orbitals (Fraction sp3; Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eC).\\u003c/p\\u003e \\u003cp\\u003eIn addition, the Tanimoto coefficient is used as an indicator to compare similarities among chemical substances. The Tanimoto coefficients of the two groups were compared (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig7\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e). DPP was used as the reference since it resulted in the highest spontaneous motor activity. Noticeable similarities were observed in the hyperactivity-associated group, with little similarity observed compared with the negative group.\\u003c/p\\u003e \\u003cp\\u003eMoreover, the chemicals in the hyperactivity-associated group were dopaminergic neurotoxins that impair the development of the dopaminergic system in the brain(Ishido, et al., \\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e2004a\\u003c/span\\u003e; \\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e2004b\\u003c/span\\u003e; \\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e2007\\u003c/span\\u003e;2017; Masuo et al., \\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e2004\\u003c/span\\u003e;). Based on this result, an investigation was conducted to determine whether the substances were classified into the two known groups of potential dopaminergic neurotoxins (Table I)(Nagatsu \\u003cspan citationid=\\\"CR32\\\" class=\\\"CitationRef\\\"\\u003e2002\\u003c/span\\u003e; Jayaraji 2016). Comparisons were performed using fingerprints similar to the Tanimoto index as an indicator. The group of dopaminergic neurotoxins containing amines was designated as DA amine, and the group of dopaminergic neurotoxins with chlorine was designated as DA oc. As shown in Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig8\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003ea, the Tanimoto coefficient in the hyperactivity-associated group was much higher than that in the other groups. As shown in Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig8\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eB, these compounds were clearly categorized into three groups by principal component analysis (PCA).\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec8\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.2. Machine learning of hyperactivity-associated chemicals\\u003c/h2\\u003e \\u003cp\\u003eMachine learning was employed to further explore the chemical nature of hyperactivity-associated chemicals. A support vector machine failed to distinguish the hyperactivity-associated components from negative components based on the relationship between spontaneous motor activity and the Tanimoto coefficient (Supplementary Fig.\\u0026nbsp;2). Neural networks succeeded in clearly distinguishing the two groups (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig9\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003eA) by gradually decreasing the loss function (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig9\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003eB); it was approximately zero at approximately 1,500 epochs (time steps; Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig9\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003eB). Thus, machine learning was applicable to our dataset.\\u003c/p\\u003e \\u003cp\\u003eFigure \\u003cspan refid=\\\"Fig10\\\" class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003e shows the random forest prediction of spontaneous motor activity (SMA) based on the Tanimoto coefficient. A good correlation was observed between the observed SMA and predicted SMA (R\\u0026thinsp;=\\u0026thinsp;0.9, mean absolute error (MAE)\\u0026thinsp;=\\u0026thinsp;0.06). The activities of two chemicals were lower than basal rat spontaneous motor activity, showing a relatively low predictive accuracy.\\u003c/p\\u003e \\u003cp\\u003eA junction tree variational autoencoder (JT-VAE) (Kingma et al. 2014; Jin et al., \\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e2019\\u003c/span\\u003e) was employed between DPP and bisphenol A to deduce the presence of the core structure in the hyperactivity-associated group,. Figure\\u0026nbsp;\\u003cspan refid=\\\"Fig11\\\" class=\\\"InternalRef\\\"\\u003e7\\u003c/span\\u003e shows the possible interpolation of two types of chemicals, suggesting the presence of a shared core structure between the two chemicals.\\u003c/p\\u003e \\u003c/div\\u003e\"},{\"header\":\"4. Discussion\",\"content\":\"\\u003cp\\u003eThere are several types of chemicals such as carcinogenic chemicals, nanomaterials, and epigenetic chemicals. In 1996, Dr. Theo Colborn \\u003cem\\u003eet al\\u003c/em\\u003e. reported oestrogenic chemicals that disturb the endocrine system in wildlife and humans, called as endocrine disruptors (Colborn et al., \\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e1996\\u003c/span\\u003e). Thus, reproductive organs should be investigated for endocrine disruptors. Due to the estrogenic activity of bisphenol A, its mode of action has been thought to be mediated by the estrogen receptor ER (Herz et al., \\u003cspan citationid=\\\"CR9\\\" class=\\\"CitationRef\\\"\\u003e2017\\u003c/span\\u003e). At present, six target molecules with which bisphenol A can interact have been identified: ER alpha, estrogen-related receptor gamma (Takayanagi et al. 2017), pregnane X receptor (Sui et al. \\u003cspan citationid=\\\"CR41\\\" class=\\\"CitationRef\\\"\\u003e2012\\u003c/span\\u003e), androgen receptor (Sohori et al., 1998), peroxisome proliferator activated receptor (Rui et al., \\u003cspan citationid=\\\"CR45\\\" class=\\\"CitationRef\\\"\\u003e2011\\u003c/span\\u003e), and thyroid hormone receptor (Moriyama et al., 2002). Thus, this fact may explain why bisphenol A induces the dysfunction of many targets, leading to the recent concept of polypharmacology (Ellingson, et al., \\u003cspan citationid=\\\"CR6\\\" class=\\\"CitationRef\\\"\\u003e2014\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eIn addition to bisphenol A, the circumstances are similar for phthalates (Harris et al., \\u003cspan citationid=\\\"CR7\\\" class=\\\"CitationRef\\\"\\u003e1997\\u003c/span\\u003e; Sun et al., \\u003cspan citationid=\\\"CR42\\\" class=\\\"CitationRef\\\"\\u003e2018\\u003c/span\\u003e). In our screening of endocrine disruptors, bisphenol A and seven phthalates exerted the same effects on inducing nervous dysfunction, suggesting that unknown similarity in both chemical structures should exist. I employed the chemoinformatics and compared two groups to elucidate this property: hyperactivity-associated chemicals and hyperactivity-negative chemicals. First, I considered what would help distinguish between hyperactivity-associated and hyperactivity-negative compounds. Using a Ward dendrogram, individual compounds were unable to be clearly distinguished based on clustering distance (Supplementary Fig.\\u0026nbsp;3). Therefore, I explored the differences in the physical properties of each group, such as molecular weight, fraction of hybridized electron orbitals (FractionCSP3), topological polar surface area, and number of rotatable bonds, and hydrogen bond donors and acceptors. Two distinctions -in FractionCSP3 and the Tanimoto coefficient were identified.\\u003c/p\\u003e \\u003cp\\u003eFraction CSP3 (Fsp3) indicates the complexity of the chemical, where Fsp3 = (number of sp3-hybridized carbons/total carbon count). Since the distribution of the molecular weight of both groups was not significantly different, the hyperactivity might be dependent on a higher number of sp3-hybridized carbons.\\u003c/p\\u003e \\u003cp\\u003eThe Tanimoto coefficient is an index of chemical similarity. The Tanimoto coefficient has two arguments: the query structure and the target structure. Dipentyl phthalate (DPP) was used as a reference since this compound resulted in the highest spontaneous motor activity (SMA) in our screen. The score of the hyperactivity-associated group was much higher than that of the negative group (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig7\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e). Although many dopaminergic toxins have been identified, as listed in Table I, hyperactivity-associated chemicals are distinguished in chemical similarity among the chemical family of dopaminergic toxins, as revealed by the Tanimoto coefficient (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig8\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eNext, I considered why the hyperactivity-associated group contained three types of functional groups (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e). I first applied machine learning to classify those data (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig9\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003e), establish a prediction (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig10\\\" class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003e), and explore the potential hidden backbone of the chemical structure; the results indicated that machine learning was applicable to those datasets.\\u003c/p\\u003e \\u003cp\\u003eWe then used a junction tree variational autoencoder (JT-VAE) (Kingma et al., 2014; Jin et al., \\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e2019\\u003c/span\\u003e). JT-VAE is a generative model that utilizes deep neural networks to describe the distribution of observed and latent (unobserved) variables. Using JT-VAE, the latent space was interpolated between DPP and bisphenol A (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig11\\\" class=\\\"InternalRef\\\"\\u003e7\\u003c/span\\u003e). Continuous interpolation seems to be performed, suggesting that a common backbone of these molecules might contribute to hyperactivity in the rats.\\u003c/p\\u003e \\u003cp\\u003eHopefully, the application of the chemical descriptors obtained in this study will facilitate chemical development to replace the hyperactivity-associated chemicals and will help elucidate the etiology of ADHD in the human.\\u003c/p\\u003e\"},{\"header\":\"5. Conclusion\",\"content\":\"\\u003cp\\u003eIn this study, we screened some endocrine disruptors based on rat hyperactivity, resulting in two groups: hyperactivity-associated group and hyperactivity-negative group. There were differences in\\u0026ldquo;Fraction sp3\\u0026rdquo; (number of sp3-hybridized carbons/total carbon count) and the Tanimoto coefficient between two groups. Our findings will help to develop more safer chemical and to elucidate the etiology of ADHD in the human.\\u003c/p\\u003e\"},{\"header\":\"Declarations\",\"content\":\"\\u003cp\\u003e\\u003cstrong\\u003eAcknowledgements\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003e\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe author thanks Drs. Ayako Furuhama, Takashi Kanamaru, Shion Honda, and Bibhash Chandra Mitra for consultations regarding chemoinformatics.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003e\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThis work was supported by a Grant-in-Aid for Scientific Research (KAKENHI).\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eCompeting interests\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe author has no conflicts of interest to declare.\\u003c/p\\u003e\"},{\"header\":\"References\",\"content\":\"\\u003cp\\u003eBen-Jonathan, N., Steinmetz, R., 1998. Xenoestrogens: the emerging story of bisphenol A. Trend Endocrinol. Metab. 9:124-128.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eBrotons, J.A., \\u003cem\\u003eet al.\\u003c/em\\u003e Xenoestrogens released from lacquer coatings in food cans. Environ. Health Perspect. 103:608-612 (1995).\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eCho, S-C. \\u003cem\\u003eet al.\\u0026nbsp;\\u003c/em\\u003e2010. Relationship between environmental phthalate exposure and the intelligence of school-age children. \\u003cem\\u003eEnviron. Health Perspect.\\u003c/em\\u003e \\u003cstrong\\u003e118,\\u0026nbsp;\\u003c/strong\\u003e1027-1032.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eChopra, V., Harley, K.,Lahiff, M., Eskenazia, B., 2014. Association between phthalates and attention deficit disorder and learning disability in U.S. children, 6\\u0026ndash;15 years. Environ Res. 128: 64\\u0026ndash;69.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eColborn, T., Dumanoski, D., Meyers, J.P., 1996. \\u003cem\\u003eOur stolen future.\\u003c/em\\u003e (Dutton, New York)\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eEllingson, S.R., Smith, J.C., Baudry, J., 2014. \\u0026nbsp;Polypharmacology and supercomputer-based docking: opportunities and challenges. \\u0026nbsp;\\u003cem\\u003eMol. Simulation\\u003c/em\\u003e \\u003cstrong\\u003e40,\\u0026nbsp;\\u003c/strong\\u003e848-854.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eHarris, C.A., Henttu, P., Parker, M.G., Sumpter, J.P., 1997. \\u0026nbsp;The estrogenic activity of phthalate esters \\u003cem\\u003ein vitro\\u003c/em\\u003e. \\u003cem\\u003eEnviron. Health Perspect.\\u003c/em\\u003e \\u003cstrong\\u003e105,\\u0026nbsp;\\u003c/strong\\u003e802-811.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eHarley, K.G. \\u003cem\\u003eet al.\\u003c/em\\u003e, 2013. Prenatal and postnatal bisphenol A and body mass index in childhood in the CHAMACOS cohort. \\u003cem\\u003eEnviron Health Perspect\\u003c/em\\u003e, 121(4): 514-520.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eHerz, C., \\u003cem\\u003eet al.,\\u0026nbsp;\\u003c/em\\u003e2017. Low-dose levels of bisphenol A inhibit telomerase via ER/GPR30-ERK signalling, impair DNA integrity and reduce cell proliferation in primary PBMC. \\u003cem\\u003eSci. Reports\\u0026nbsp;\\u003c/em\\u003e7, doi:10.1038/s41598-017-15978-2.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eHonda, S., 2019\\u003cem\\u003e\\u0026nbsp;Interpolation with pre-trained JTNN\\u003c/em\\u003e.\\u0026nbsp;https:// github.com/ shionhonda/ dgl-playground/blob/master/jtnn.ipynb. (accessed 6 April 2022)\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eHuang, H-B. \\u003cem\\u003eet al.\\u003c/em\\u003e, 2015.Fetal and Childhood Exposure to Phthalate Diesters and Cognitive Function in Children Up to 12 Years of Age: Taiwanese Maternal and Infant Cohort Study. \\u003cem\\u003ePLOS One\\u003c/em\\u003e \\u003ca href=\\\"https://doi.org/10.1371/journal.pone.0131910\\\"\\u003ehttps://doi.org/10.1371/journal.pone.0131910\\u003c/a\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eIshido, M., Masuo, Y., Oka, S., Kunimoto, M., Morita, M., 2002. \\u0026nbsp;Application of Supermex system to screen behavioral traits produced by tributyltin in the rat. \\u003cem\\u003eJ. Health Sci.\\u003c/em\\u003e \\u003cstrong\\u003e48,\\u003c/strong\\u003e 451-454.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eIshido, M., Masuo, Y., Kunimoto, M., Oka, S., Morita, M., 2004a. Bisphenol A causes hyperactivity in the rat concomitantly with impairment of tyrosine hydroxylase immunoreactivity. \\u0026nbsp;\\u003cem\\u003eJ. Neurosci. Res.\\u003c/em\\u003e \\u003cstrong\\u003e76,\\u003c/strong\\u003e 423-433. \\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eIshido, M., Masuo, Y., Oka S, Niki E, Morita, M. 2004b. \\u003cem\\u003ep\\u003c/em\\u003e-Nitrotoluene causes hyperactivity in the rat. \\u003cem\\u003eNeurosci. Lett\\u003c/em\\u003e. \\u003cstrong\\u003e366,\\u003c/strong\\u003e 1-5.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eIshido, M.,\\u003cem\\u003e\\u0026nbsp;et al.,\\u003c/em\\u003e 2004c.\\u003cem\\u003e\\u0026nbsp;\\u003c/em\\u003eDicyclohexylphthalate causes hyperactivity in the rat concomitantly with impairment of tyrosine hydroxylase immunoreactivity, \\u003cem\\u003eJ. Neurochem.\\u003c/em\\u003e \\u003cstrong\\u003e91,\\u003c/strong\\u003e 69-76.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eIshido, M., Morita, M., Oka, S., Masuo Y., 2005.Alteration of gene expression of G-protein-coupled receptors in endocrine disruptors-caused hyperactive rats. \\u003cem\\u003eRegulatory Peptides\\u0026nbsp;\\u003c/em\\u003e\\u003cstrong\\u003e126,\\u003c/strong\\u003e 145-153.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eIshido, M., Yonemoto, J., Morita, M., 2007. Mesencephalic neurodegeneration in the orally administered bisphenol A-caused hyperactive rats. \\u003cem\\u003eToxicol. Lett\\u003c/em\\u003e. \\u003cstrong\\u003e173,\\u003c/strong\\u003e 66-72. \\u0026nbsp;\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eIshido, M., Masuo, Y., Terasaki, M., Morita, M., 2011. Rat hyperactivity by bisphenol A, but not by its derivatives, 3-hydroxybisphenol A or bisphenol A 3,4-quinone. \\u003cem\\u003eToxicol. Lett.\\u003c/em\\u003e \\u003cstrong\\u003e206,\\u003c/strong\\u003e 300-305. \\u0026nbsp;\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eIshido, M., Usu, R., 2017. Orally administered \\u003cem\\u003ep\\u003c/em\\u003e-nitrotoluene causes hyperactivity, concomitantly with gliosis and impairment of tyrosine hydroxylase immunoreactivity in the rat substantia nigra. \\u003cem\\u003eFund. Toxicol. Sci.\\u003c/em\\u003e \\u003cstrong\\u003e4,\\u0026nbsp;\\u003c/strong\\u003e151-158.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eJacobson, J.L., Jacobson, S.W., 1996. Intellectual impairment in children exposed to polychlorinated biphenyls in utero. \\u0026nbsp; \\u003cem\\u003eNew Engl. J. Med.\\u003c/em\\u003e 335:783-789.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eJayaraj, R., Megha, P., Sreedev, P., 2016. Organochlorine pesticides, their toxic effects on living organisms and their fate in the environment. \\u003cem\\u003eInterdiscip. Toxicol.\\u003c/em\\u003e \\u003cstrong\\u003e9,\\u0026nbsp;\\u003c/strong\\u003e90-100.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eJin, W., Barzilay, R., Jaakkola, T., 2019. Junctional tree variational autoencoder for molecular graph generation. \\u003cem\\u003earXiv Prepr. arXiv: 1802.04364v4\\u003c/em\\u003e.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eKanamaru, T., 2018. \\u003cem\\u003eMachine learning with Raspberry Pi.\\u0026nbsp;\\u003c/em\\u003e(Kodansha, Tokyo) (\\u003cstrong\\u003ein Japanese\\u003c/strong\\u003e)\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eKingma, D., Welling, M., 2014. Auto-encoding variational bays. \\u003cem\\u003earXiv Prepr. arXiv1312.6114.\\u003c/em\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eKim, B-N. \\u003cem\\u003eet al.\\u003c/em\\u003e, 2009. Phthalates exposure and attention-deficit/hyperactivity disorder in school-age children. \\u003cem\\u003eBiol Psychiatry\\u003c/em\\u003e \\u003cstrong\\u003e66,\\u003c/strong\\u003e 958-63.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eLondon, F., Etzel, R.A., 2000. The environment as an etiologic factor in autism: a new direction for research. Environ. Health Persp. 108 (Suppl 3):401-404.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eMasuo Y, Matsumoto Y, Morita S, Noguchi J., 1997. A novel method for counting spontaneous motor activity in the rat. \\u003cem\\u003eBrain Res, Protocols\\u003c/em\\u003e \\u003cstrong\\u003e1,\\u003c/strong\\u003e 321-326.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eMasuo, Y., Ishido, M., Morita, M., Oka, S., 2004. Effects of neonatal treatment with 6-hydroxydopamine and endocrine disruptors on motor activity and gene expression in rats. \\u003cem\\u003eNeural Plasticity\\u0026nbsp;\\u003c/em\\u003e\\u003cstrong\\u003e11,\\u0026nbsp;\\u003c/strong\\u003e59-76.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eMiodovnik, A. \\u003cem\\u003eet al.,\\u003c/em\\u003e 2011. Endocrine disruptors and childhood social impairment. \\u003cem\\u003eNeurotoxicology.\\u003c/em\\u003e \\u003cstrong\\u003e32,\\u0026nbsp;\\u003c/strong\\u003e261\\u0026ndash;267.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eMori, C., 2001. Assessment of fetal environmental exposure and subtle male reproductive disorders in Japan. \\u0026nbsp;In Proceedings of the 1st U.S.-Japan workshop on endocrine disrupting chemicals and their toxicological evaluation. 103-106.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eMoriyama, K. \\u003cem\\u003eet. al.\\u003c/em\\u003e, 2002. Thyroid hormone action is disrupted by bisphenol A as an antagonist. \\u003cem\\u003eJ. Clin. Endocrinol. Metab.\\u003c/em\\u003e \\u003cstrong\\u003e87,\\u003c/strong\\u003e 5185-5190.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eNagatsu, T., 2002. Amine-related neurotoxins in Parkinson\\u0026rsquo;s disease: Past, present, and future. \\u003cem\\u003eNeurotoxicol. Teratology\\u003c/em\\u003e \\u003cstrong\\u003e24,\\u0026nbsp;\\u003c/strong\\u003e565-569.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eNTP (National Toxicology Program), 1998. \\u003cem\\u003eReport on carcinogens\\u003c/em\\u003e, eighth edition. National Toxicology Program: National Institutes of Health (NIH) publications, Research Triangle, Park, NC, USA.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003ePark, S. \\u003cem\\u003eet al.\\u003c/em\\u003e, 2015. Association between phthalates and externalizing behaviors and cortical thickness in children with attention deficit hyperactivity disorder. \\u003cem\\u003ePsychol Med\\u003c/em\\u003e \\u003cstrong\\u003e45,\\u003c/strong\\u003e 1601-12. doi: 10.1017/S0033291714002694.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003ePatandin, S. et al., 1994. Effects of environmental exposure to polychlorinated biphenyls and dioxins on cognitive abilities in Dutch children at 42 months of age. \\u0026nbsp;J. Pediatr 134\\u003cstrong\\u003e:\\u003c/strong\\u003e33-41.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003ePolanska, K., Ligocka, D., Sobala, W., Hanke, W., 2014. Phthalate exposure and child development: the Polish Mother and Child Cohort Study. \\u003cem\\u003eEarly Hum. Dev.\\u003c/em\\u003e \\u003cstrong\\u003e90,\\u003c/strong\\u003e 477-485.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eRice, D.C., 2000. Parallels between attention deficit hyperactivity disorder and behavioral deficits produced by neurotoxic exposure in monkeys. \\u0026nbsp;Environ. Health Persp. 108(Suppl 3):405-408.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eShaywitz, B.A., Yager, R.D., Klopper, J.H., 1976. Selective brain dopamine depletion in developing rats: an experimental model of minimal brain dysfunction. Science 191, 305-308.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eSohoni, P., Sumpter, J.P., 1998. Several environmental oestrogens are also anti-androgens. \\u0026nbsp;\\u003cem\\u003eJ. Endocrinol.\\u0026nbsp;\\u003c/em\\u003e\\u003cstrong\\u003e158,\\u003c/strong\\u003e 327-339.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eStein, P., \\u003cem\\u003eet al.\\u003c/em\\u003e, 2015. Bisphenol A exposure in children with autism spectrum disorders. \\u003cem\\u003eAutism Res.\\u003c/em\\u003e \\u003cstrong\\u003e8,\\u003c/strong\\u003e 272-83. doi: 10.1002/aur.1444.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eSui, Y. \\u003cem\\u003eet al.\\u003c/em\\u003e, 2012. Bisphenol A and its analogues activate human pregnane X receptor. \\u003cem\\u003eEnviron. Health Perspect.\\u003c/em\\u003e \\u003cstrong\\u003e120,\\u0026nbsp;\\u003c/strong\\u003e399-405.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eSun, D. \\u003cem\\u003eet al.\\u003c/em\\u003e, 2018.\\u003cem\\u003e\\u0026nbsp;\\u003c/em\\u003eEffect of di-(2-ethylhexyl)phthalate on the hypothalamus-pituitary-thyroid axis in adolescent rat. \\u003cem\\u003eEndocrine J.\\u003c/em\\u003e \\u003cstrong\\u003e65,\\u0026nbsp;\\u003c/strong\\u003e261-268.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eTakayanagi, S. \\u003cem\\u003eet al.\\u003c/em\\u003e, 2006. Endocrine disruptor bisphenol A strongly binds to human estrogen-related receptor gamma (ERR gamma) with high constitutive activity. \\u003cem\\u003eToxicol. Lett.\\u003c/em\\u003e \\u003cstrong\\u003e167,\\u0026nbsp;\\u003c/strong\\u003e95-105.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eTewar, S. \\u003cem\\u003eet al.\\u003c/em\\u003e, 2016. Association of Bisphenol A exposure and Attention-Deficit/Hyperactivity Disorder in a national sample of U.S. children. \\u003cem\\u003eEnviron. Res.\\u0026nbsp;\\u003c/em\\u003e150:112-118. doi: 10.1016/j.envres.2016.05.040.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eRui, A. \\u003cem\\u003eet al.,\\u003c/em\\u003e 2011. Peroxisome proliferator-activated receptor gamma is a target for halogenated analogs of bisphenol A. \\u0026nbsp;\\u003cem\\u003eEnviron. Health Perspect.\\u003c/em\\u003e \\u003cstrong\\u003e119,\\u0026nbsp;\\u003c/strong\\u003e1227-1232.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eYolton, K. \\u003cem\\u003eet al.\\u003c/em\\u003e, 2011. Prenatal exposure to bisphenol A and phthalates and infant neurobehavior. \\u003cem\\u003eNeurotoxicol. Teratol.\\u003c/em\\u003e\\u003cstrong\\u003e33,\\u0026nbsp;\\u003c/strong\\u003e558-66. \\u003c/p\\u003e\"},{\"header\":\"Table\",\"content\":\"\\u003cp\\u003eTable \\u003cu\\u003eI\\u003c/u\\u003e. Potential dopaminergic toxins\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003e\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003ctable border=\\\"1\\\" cellpadding=\\\"0\\\" cellspacing=\\\"0\\\"\\u003e\\n \\u003ctbody\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd valign=\\\"top\\\" width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003eChemicals\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd valign=\\\"top\\\" width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003e\\u003cem\\u003eAmine\\u003c/em\\u003e\\u003cem\\u003e \\u003c/em\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd valign=\\\"top\\\" width=\\\"100%\\\"\\u003e\\n \\u003ctable border=\\\"0\\\" cellpadding=\\\"0\\\" cellspacing=\\\"0\\\" width=\\\"406\\\"\\u003e\\n \\u003ctbody\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003eMPP+\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003eTaClo-d4\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003eNorharman(\\u0026beta;Carboline)\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003e1,2,3,4-Tetrahydroisoquinoline\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003e1-Phenyl-1,2,3,4-tetrahydroisoquinoline\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003eSalsolinol（1-Methyl-6,7-dihydroxytetrahydroisoquinoline）\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003eNorsalsolinol\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003c/tbody\\u003e\\n \\u003c/table\\u003e\\n \\u003cp\\u003e\\u003cbr\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd valign=\\\"top\\\" width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003e\\u003cem\\u003eOrganochlorines\\u003c/em\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd valign=\\\"top\\\" width=\\\"100%\\\"\\u003e\\n \\u003ctable border=\\\"0\\\" cellpadding=\\\"0\\\" cellspacing=\\\"0\\\" width=\\\"403\\\"\\u003e\\n \\u003ctbody\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003eDDT\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003eLindane\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003eDieldrin\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003eHeptachlor\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003eMethoxychlor\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003eTrichloroethylene\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003eAldrin\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003eChlordane\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd width=\\\"100%\\\"\\u003e\\n \\u003cp\\u003eEndosulfan\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003c/tbody\\u003e\\n \\u003c/table\\u003e\\n \\u003cp\\u003e\\u003cbr\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003c/tbody\\u003e\\n\\u003c/table\\u003e\\n\\u003cp\\u003e\\u0026nbsp;\\u003c/p\\u003e\"}],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":true,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":true,\"hideJournal\":true,\"highlight\":\"\",\"institution\":\"National Institute for Environmental Studies\",\"isAcceptedByJournal\":false,\"isAuthorSuppliedPdf\":false,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":false,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"researchsquare\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":true,\"externalIdentity\":\"\",\"sideBox\":\"\",\"snPcode\":\"\",\"submissionUrl\":\"/submission\",\"title\":\"Research Square\",\"twitterHandle\":\"researchsquare\",\"acdcEnabled\":true,\"dfaEnabled\":false,\"editorialSystem\":\"\",\"reportingPortfolio\":\"\",\"inReviewEnabled\":false,\"inReviewRevisionsEnabled\":true},\"keywords\":\"Environmental chemicals, ADHD, Toxicoinformatics, Chemoinformatics, Machine learning \",\"lastPublishedDoi\":\"10.21203/rs.3.rs-1701860/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-1701860/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"\\u003cp\\u003e・Using the bioassay based on rat hyperactivity, some endocrine disruptors were categorized into hyperactivity-associated group and hyperactivity-negative group.\\u003c/p\\u003e \\u003cp\\u003e・Two groups have differences in\\u0026ldquo;Fraction sp3\\u0026rdquo; (number of sp3-hybridized carbons/total carbon count) and the Tanimoto coefficient.\\u003c/p\\u003e \\u003cp\\u003e・A neural network model clearly classified the two groups. Random forest methods also showed the good prediction (R\\u0026thinsp;=\\u0026thinsp;0.9, MAE (mean absolute error)\\u0026thinsp;=\\u0026thinsp;0.06).\\u003c/p\\u003e \\u003cp\\u003e・Using a junction tree variational autoencoder, the core structure was interpolated between phthalate and phenol in the hyperactivity-associated group.\\u003c/p\\u003e\",\"manuscriptTitle\":\"Chemical nature of attention deficit hyperactivity disorder (ADHD)- related chemical subfamily.\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2022-05-31 18:28:16\",\"doi\":\"10.21203/rs.3.rs-1701860/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"researchsquare\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":true,\"externalIdentity\":\"\",\"sideBox\":\"\",\"snPcode\":\"\",\"submissionUrl\":\"/submission\",\"title\":\"Research Square\",\"twitterHandle\":\"researchsquare\",\"acdcEnabled\":true,\"dfaEnabled\":false,\"editorialSystem\":\"\",\"reportingPortfolio\":\"\",\"inReviewEnabled\":false,\"inReviewRevisionsEnabled\":true}}],\"origin\":\"\",\"ownerIdentity\":\"85f976bb-5c69-42e2-b1d1-058bdbc6f3fd\",\"owner\":[],\"postedDate\":\"May 31st, 2022\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"posted\",\"subjectAreas\":[{\"id\":12894086,\"name\":\"Toxicology\"}],\"tags\":[],\"updatedAt\":\"2022-05-31T18:28:16+00:00\",\"versionOfRecord\":[],\"versionCreatedAt\":\"2022-05-31 18:28:16\",\"video\":\"\",\"vorDoi\":\"\",\"vorDoiUrl\":\"\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-1701860\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-1701860\",\"identity\":\"rs-1701860\",\"version\":[\"v1\"]},\"buildId\":\"FbvkV6FR0MCFSLy54lSbu\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}