Virtual Screening Tool Research Articles

A virtual reality cognitive health screening tool for aviation: Managing accident risk for older pilots

To address elevated risk for older pilots, we examined the efficacy of a virtual reality (VR) cognitive health screening tool (integrated into simulated flight scenarios) in identifying general aviation pilots who experienced a critical incident during flight in a full-scale Cessna 172 simulator. Performance data were obtained from 51 certified pilots (17–71 years). Machine learning classification algorithms, based on key data from the VR flight, were used to validate the utility of the screening tool for identifying pilot risk. The results showed that aviation-relevant cognitive factors obtained in the VR screening tool, including situation awareness and prospective memory, predicted risk of a critical incident with good sensitivity (0.83) and specificity (0.85), AUC = 0.82. These results support VR-based cognitive screening to identify at-risk older pilots. The present findings inform procedures for optimizing safety and reducing critical incidents at any point in the pilot lifespan and are timely in view of the impending pilot workforce shortage.

A comprehensive comparative assessment of 3D molecular similarity tools in ligand-based virtual screening.

Three-dimensional (3D) molecular similarity, one major ligand-based virtual screening (VS) method, has been widely used in the drug discovery process. A variety of 3D molecular similarity tools have been developed in recent decades. In this study, we assessed a panel of 15 3D molecular similarity programs against the DUD-E and LIT-PCBA datasets, including commercial ROCS and Phase, in terms of screening power and scaffold-hopping power. The results revealed that (1) SHAFTS, LS-align, Phase Shape_Pharm and LIGSIFT showed the best VS capability in terms of screening power. Some 3D similarity tools available to academia can yield relatively better VS performance than commercial ROCS and Phase software. (2) Current 3D similarity VS tools exhibit a considerable ability to capture actives with new chemotypes in terms of scaffold hopping. (3) Multiple conformers relative to single conformations will generally improve VS performance for most 3D similarity tools, with marginal improvement observed in area under the receiving operator characteristic curve values, enrichment factor in the top 1% and hit rate in the top 1% values showed larger improvement. Moreover, redundancy and complementarity analyses of hit lists from different query seeds and different 3D similarity VS tools showed that the combination of different query seeds and/or different 3D similarity tools in VS campaigns retrieved more (and more diverse) active molecules. These findings provide useful information for guiding choices of the optimal 3D molecular similarity tools for VS practices and designing possible combination strategies to discover more diverse active compounds.

Virtual screening of potential phyto-candidates as therapeutic leads against SARS-CoV-2 infection

The outbreak of novel coronavirus strain (Covid-19) with a high pandemic threat has predict grave public health and economic concerns. This virus, originating from the Wuhan region in China has spread worldwide affecting millions with no registered persuasive targeted therapy. In this paper, we analyze the three important proteins encoded by the virus, envelope protein 5 × 29, RNA binding nucleocapsid protein 1SSK, and spike glycoprotein 6ACD, for an effective virion accumulation, and remdesivir was the first drug approved by the FDA and EMA for the treatment of COVID-19 cases that require hospitalization, there is still much controversy about its efficacy and also an alternative for novel phytochemicals, deoxynojirimycin, trigoneoside IB, and octanoic acid. The in-silico evaluations were conducted using the PyRx virtual screening tools which lead to the target based on high binding affinity. Trigoneoside IB, derived from Trigonella foenum-graecum (Fenugreek), showed the highest binding affinity and stable interaction with the amino acid residues present in active sites of Covid-19 proteins. Meanwhile, the other two compounds derived from Morus alba (Mulberry) and Morinda citrifolia (Noni), as well as the anti-HIV remdesivir drug exhibited good binding affinity and favorable ADME properties. Thereby offering scope for validation of the new therapeutic components for their in vitro and in vivo efficacy against the Covid-19 proteins.

Cannabichromene and Δ9-Tetrahydrocannabinolic Acid Identified as Lactate Dehydrogenase-A Inhibitors by in Silico and in Vitro Screening.

Cannabis sativa contains >120 phytocannabinoids, but our understanding of these compounds is limited. Determining the molecular modes of action of the phytocannabinoids may assist in their therapeutic development. Ligand-based virtual screening was used to suggest novel protein targets for phytocannabinoids. The similarity ensemble approach, a virtual screening tool, was applied to target identification for the phytocannabinoids as a class and predicted a possible interaction with the lactate dehydrogenase (LDH) family of enzymes. In order to evaluate this in silico prediction, a panel of 18 phytocannabinoids was screened against two LDH isozymes (LDHA and LDHB) in vitro. Cannabichromene (CBC) and Δ9-tetrahydrocannabinolic acid (Δ9-THCA) inhibited LDHA via a noncompetitive mode of inhibition with respect to pyruvate, with Ki values of 8.5 and 6.5 μM, respectively. In silico modeling was then used to predict the binding site for CBC and Δ9-THCA. Both were proposed to bind within the nicotinamide pocket, overlapping the binding site of the cofactor NADH, which is consistent with the noncompetitive modes of inhibition. Stemming from our in silico screen, CBC and Δ9-THCA were identified as inhibitors of LDHA, a novel molecular target that may contribute to their therapeutic effects.

In Silico Identification of Flavonoids from Corriandrum sativum Seeds against Coronavirus Covid-19 Main Protease

Molecular docking analysis is routinely used in modern drug research to understand and predict the relationship between a drug molecule and a target protein from a microbe. The entry and replication of pathogens in host cells can be prevented by drugs identified in this way. The coronavirus disease associated with SARS-CoV-2, COVID-19, has become today's most infectious and lethal pandemic disease in the world. Burgeoning in the absence of any particular vaccine or therapeutic agent against SARS-CoV-2.The situation urges the need for appropriate medications to treat patients infected with the virus. Consequently, the study focus on evaluate the therapeutic potential of flavonoids present in Corriandrum sativum seeds that could serve as suitable remedies for COVID19.We analyzed the binding affinity of four flavonoids were screened against Mpro protein of SARS-CoV-2 by PyRx Virtual Screening tool and also results are validated with Lig-Plot Plus. Lopinavir shows binding affinity of -8.3 Kcal/mol and exhibit stable, strong interaction with active site of COVID19 main protease. Besides flavonoids, Rutin found to have the highest binding affinity compared to Lopinavir with the Mpro protease, followed by Chlorogenic acid, Quercetin and Caffeic acid. The present study concludes that Rutin present in the integrant of seeds shows the highest potentiality for acting as in inhibitor of main protease enzyme. Further, characterization of the amino acid residues comprising the viral binding site and the nature of the hydrogen bonding involved in the ligand receptor interaction shows significant findings with Rutin binding to the MPro protein at amino acid. The amino acid acid present in active sites of Mpro protease responsible for virus pathogenicity. The findings of the present study need in vivo experiments to prove the utility of Rutin compounds and further use in making Corriandrum sativum seeds as anti-SARS-CoV-2 product in near future.
 Keywords: Corriandrum sativum seeds,Novel Coronavirus, SARS-CoV2, COVID-19, Protease, Molecular Docking.

Virtual Cocrystal Screening Methods as Tools to Understand the Formation of Pharmaceutical Cocrystals—A Case Study of Linezolid, a Wide-Range Antibacterial Drug

Experimental mechanochemical screening of cocrystals with linezolid (LIN) resulted in the formation of six new crystal phases, including three neat cocrystals and three cocrystal hydrates, in addition to seven previously described cocrystals. In an attempt to understand the factors governing the formation of these phases, different experimental conditions of the mechanochemical reactions (polymorphic forms of LIN and presence of different solvents to create liquid-assisted grinding conditions) were tested and the results were compared with the predictions from three commonly used virtual cocrystal screening tools: molecular complementarity, hydrogen bond propensity, and molecular electrostatic potential maps. It is shown that these three methods can be used to help understand a molecule’s preferences to form cocrystals with particular coformers. The influence of molecular conformation on the outcome of the predictions is also evaluated. A comparison between the prediction methods indicates that while considering a set of similar coformers, the approach based on molecular electrostatic potential maps seems to be more consistent with the experimental results than molecular complementarity and hydrogen bond propensity tools. Instead, these two latter approaches are recommended at the early stages of coformer selection. In addition, intermolecular energy contribution (lattice energy) to the total energy of crystal forms of coformers was found to be indicative of the feasibility of cocrystal formation in the case of coformers capable of forming similar supramolecular synthons.

General Purpose Structure-Based Drug Discovery Neural Network Score Functions with Human-Interpretable Pharmacophore Maps.

The BioChemical Library (BCL) is an academic open-source cheminformatics toolkit comprising ligand-based virtual high-throughput screening (vHTS) tools such as quantitative structure–activity/property relationship (QSAR/QSPR) modeling, small molecule flexible alignment, small molecule conformer generation, and more. Here, we expand the capabilities of the BCL to include structure-based virtual screening. We introduce two new score functions, BCL-AffinityNet and BCL-DockANNScore, based on novel distance-dependent signed protein–ligand atomic property correlations. Both metrics are conventional feed-forward dropout neural networks trained on the new descriptors. We demonstrate that BCL-AffinityNet is one of the top performing score functions on the comparative assessment of score functions 2016 affinity prediction and affinity ranking tasks. We also demonstrate that BCL-AffinityNet performs well on the CSAR-NRC HiQ I and II test sets. Furthermore, we demonstrate that BCL-DockANNScore is competitive with multiple state-of-the-art methods on the docking power and screening power tasks. Finally, we show how our models can be decomposed into human-interpretable pharmacophore maps to aid in hit/lead optimization. Altogether, our results expand the utility of the BCL for structure-based scoring to aid small molecule discovery and design. BCL-AffinityNet, BCL-DockANNScore, and the pharmacophore mapping application, as well as the remainder of the BCL cheminformatics toolkit, are freely available with an academic license at the BCL Commons site hosted on http://meilerlab.org/.

A Computational Approach on the Activity of Hesperidin as Antagonist for Proteins of SARS-CoV-2

Spike proteins are a functional component in the viral infections of the SARS-CoV-2 virus that binds to the ACE2 receptors of the human cells. The naturally derived drugs like hesperidin show a higher affinity to avoiding the binding of the spike protein. Inhibition of the spike protein study represents the development of a new drug for SARS disease. Along with the hesperidin and its analogs have been found effective medicine to control SARS-CoV-2; the entire drug bank database was screened for strong analogous remedy like compounds as hesperidin. Virtual Screening and docking studies were intended for these molecules against spike protein with auto dock virtual screening tool and auto dock vena. The docking outcome showed that the compounds hesperitin, silibilin, and dihydromyricetin were having the highest binding energies, like -8 and -6.2. The current study indicates that the lead molecules have to be evaluated to improve prospective drug molecules.

WIDOCK: a reactive docking protocol for virtual screening of covalent inhibitors

Here we present WIDOCK, a virtual screening protocol that supports the selection of diverse electrophiles as covalent inhibitors by incorporating ligand reactivity towards cysteine residues into AutoDock4. WIDOCK applies the reactive docking method (Backus et al. in Nature 534:570–574, 2016) and extends it into a virtual screening tool by introducing facile experimental or computational parametrization and a ligand focused evaluation scheme together with a retrospective and prospective validation against various therapeutically relevant targets. Parameters accounting for ligand reactivity are derived from experimental reaction kinetic data or alternatively from computed reaction barriers. The performance of this docking protocol was first evaluated by investigating compound series with diverse warhead chemotypes against KRASG12C, MurA and cathepsin B. In addition, WIDOCK was challenged on larger electrophilic libraries screened against OTUB2 and NUDT7. These retrospective analyses showed high sensitivity in retrieving experimental actives, by also leading to superior ROC curves, AUC values and better enrichments than the standard covalent docking tool available in AutoDock4 when compound collections with diverse warheads were investigated. Finally, we applied WIDOCK for the prospective identification of covalent human MAO-A inhibitors acting via a new mechanism by binding to Cys323. The inhibitory activity of several predicted compounds was experimentally confirmed and the labelling of Cys323 was proved by subsequent MS/MS measurements. These findings demonstrate the usefulness of WIDOCK as a warhead-sensitive, covalent virtual screening protocol.

A GPU based virtual screening tool using SOM

A GPU based virtual screening tool using SOM

A GPU based virtual screening tool using SOM

This paper attempts to introduce the applicability of low cost graphics processing unit alternatives to a virtual screening technique using a novel self-organising map (SOM) based technique. This method combines the unsupervised learning capability of the SOM with a subsequent supervised labelling of the trained SOM neurons for building the prediction model. This novel iteration-based SOM technique can label molecule as undefined classes which can reduce the false positives in the screening. For running large datasets, the serial implementation of the proposed algorithm is very time-consuming and cannot be completed in a stipulated time frame. This has been overcome by exploiting the parallelism present in finding the winner neuron and neuron weight updating steps. A tool named SOMSCREEN is developed based on the proposed parallelised method to make the drug discovery process faster. It is observed that, the proposed method offers reduced false positive rate than the Random Forest based work.

In silico Design of Novel N-hydrosulfonylbenzamides inhibitors of dengue RNA-dependent RNA polymerase showing favorable predicted pharmacokinetic profiles

Background: In recent years, there has been a growing interest in Denv NS5 inhibition, with several reported RdRp inhibitors such as sulfonylbenzamides, non-nucleo-side inhibitors without any 3D-QSAR pharmacophore (PH4) available. In this context, we report here, in silico design and virtual evaluation of novel sulfonylbenzamides Denv RdRp inhibitors with favorable predicted pharmacokinetic profile. Methods: By using in situ modifications of the crystal structure of 5-(5-(3-hydroxyprop-1-yn-1-yl)thiophen-2-yl)-4- methoxy-2-methyl-N-(methylsulfonyl) benzamide (EHB)-RdRp complex (PDB entry 5HMZ), 3D models of RdRp-EHBx complexes were prepared for a training set of 18 EHBs with experimentally determined inhibitory potencies (half-maximal inhibitory concentrations IC50exp). In the search for active conformation of the EHB1-18, linear QSAR model was prepared, which correlated computed gas phase enthalpies of formation ∆∆HMM of RdRp-EHBx complexes with the IC50exp. Further, considering the solvent effect and entropy changes upon ligand binding resulted in a superior QSAR model correlating computed complexation Gibbs free energies (∆∆Gcom). The successive pharmacophore model (PH4) generated from the active conformations of EHBs served as a virtual screening tool of novel analogs included in a virtual combinatorial library (VCL) of compounds with scaffolds restricted to phenyl. The VCL filtered by the Lipinski’s rule-of-five was screened by the PH4 model to identify new EHB analogs. Results: Gas phase QSAR model: -log10(IC50exp) = p IC50exp =-0.1403 x ∆∆HMM _ 7.0879, R2 = 0.73; superior aqueous phase QSAR model: p IC50exp = -0.2036 x ∆∆Gcom + 7.4974, R2 = 0.81 and PH4 pharmacophore model: p IC50exp = 1.0001 x p IC50pre -0.0017, R2 = 0.97. The VCL of more than 30 million EHBs was filtered down to 125,915 analogs Lipinski’s rule. The five-point PH4 screening retained 329 new and potent EHBs with predicted inhibitory potencies p IC50pre up to 30 times lower than that of EHB1 (IC50exp = 23nM). Predicted pharmacokinetic profile of the new analogs showed enhanced cell membrane permeability and high human oral absorption compared to the alone drug to treat dengue virus. Conclusions: Combined use of QSAR models, which considered binding of the EHBs to RdRp, pharmacophore model and ADME properties helped to recognize bound active conformation of the sulfonylbenzamide inhibitors, permitted in silico screening of VCL of compounds sharing sulfonylbenzamide scaffold and identify new analogs with predicted high inhibitory potencies and favorable pharmacokinetic profiles. Keywords: ADME properties prediction, Dengue, 3-(5-ethynylthiophen-2-yl)-N-hydrosulfonylbenzamides, in silico screening, RNA-dependent RNA polymerase.

In silico exploration of novel protease inhibitors against coronavirus 2019 (COVID-19).

The spread of SARS-CoV-2 has affected human health globally. Hence, it is necessary to rapidly find the drug-candidates that can be used to treat the infection. Since the main protease (Mpro) is the key protein in the virus's life cycle, Mpro is served as one of the critical targets of antiviral treatment. We employed virtual screening tools to search for new inhibitors to accelerate the drug discovery process. The hit compounds were subsequently docked into the active site of SARS-CoV-2 main protease and ranked by their binding energy. Furthermore, in-silico ADME studies were performed to probe for adoption with the standard ranges. Finally, molecular dynamics simulations were applied to study the protein-drug complex's fluctuation over time in an aqueous medium. This study indicates that the interaction energy of the top ten retrieved compounds with COVID-19 main protease is much higher than the interaction energy of some currently in use protease drugs such as ML188, nelfinavir, lopinavir, ritonavir, and α-ketoamide. Among the discovered compounds, Pubchem44326934 showed druglike properties and was further analyzed by MD and MM/PBSA approaches. Besides, the constant binding free energy over MD trajectories suggests a probable drug possessing antiviral properties. MD simulations demonstrate that GLU166 and GLN189 are the most important residues of Mpro, which interact with inhibitors.

Virtual screening and repurposing of FDA-approved drugs from ZINC database to identify potential autophagy inhibitors exploiting autophagy related 4A cysteine peptidase as a target: potential as novel anti-cancer molecule.

Cancer cells utilize extensive autophagy in effort to adapt to high metabolic stress. This indicates that impairing the high autophagic flux might be an attractive target for cancer therapy. Autophagy related gene 4A (ATG4A) is a key player for autophagy and its inhibition may help in tumor clearance. The present study aims to screen candidate drugs from FDA-approved drugs, a subset of Zinc database, to identify potential ATG4A inhibitors that may have anti-cancer activity. Computer aided drug design approach was applied for the study using the virtual screening tools Raccoon and MGLTools-1.5.6. We have identified the drug Lumacaftor as a potent inhibitor of ATG4A on the basis of computational approaches viz. molecular docking, molecular dynamics simulation and MM/PBSA method. The drug is likely to be a potent regimen candidate to be used as an anti-cancer molecule. However, this potent inhibitor against ATG4A as anti-cancer molecule needs further investigation and validation. Communicated by Ramaswamy H. Sarma

Fish tyrosinase enzyme involved in melanin biosynthesis: Insights from physicochemical characterization, homology modeling, and virtual screening studies

In the vertebrates, including fish, the tyrosinase enzyme plays an essential role in coloration. Modulation of tyrosinase activity is expected to alter the body pigmentation in fish and other vertebrate species. In the present study, physicochemical, functional, and structural properties of tyrosinase of three fish species viz., goldfish, Japanese medaka, and common carp were determined. The homology model was developed using the Chimera1.1.2, Swiss model, and Phyre2, and the best model was selected upon evaluation. Further, a virtual screening method was applied to identify the putative modulators using the PyRx- Virtual screening tool. The estimated physicochemical and functional properties of tyrosinase from the three species suggested that they all are hydrophobic, acidic, thermostable, with a high extinction coefficient (Cys, Trp, and Tyr) and have transmembrane-segment. Based on virtual screening against 13,000 compounds from the zinc database, five compounds were determined as potent modulators of fish tyrosinase with a binding energy of −7.0 to −8.8 Kcal/mol. Of these, Pilosine (ZINC13469966) was found to be the best putative modulator with low binding energy and properties of standardized drugs. This study showed that the tyrosinase function could be modulated to alter the pigment formation in fish species by using small compound.

Seq-SymRF: a random forest model predicts potential miRNA-disease associations based on information of sequences and clinical symptoms

Increasing evidence indicates that miRNAs play a vital role in biological processes and are closely related to various human diseases. Research on miRNA-disease associations is helpful not only for disease prevention, diagnosis and treatment, but also for new drug identification and lead compound discovery. A novel sequence- and symptom-based random forest algorithm model (Seq-SymRF) was developed to identify potential associations between miRNA and disease. Features derived from sequence information and clinical symptoms were utilized to characterize miRNA and disease, respectively. Moreover, the clustering method by calculating the Euclidean distance was adopted to construct reliable negative samples. Based on the fivefold cross-validation, Seq-SymRF achieved the accuracy of 98.00%, specificity of 99.43%, sensitivity of 96.58%, precision of 99.40% and Matthews correlation coefficient of 0.9604, respectively. The areas under the receiver operating characteristic curve and precision recall curve were 0.9967 and 0.9975, respectively. Additionally, case studies were implemented with leukemia, breast neoplasms and hsa-mir-21. Most of the top-25 predicted disease-related miRNAs (19/25 for leukemia; 20/25 for breast neoplasms) and 15 of top-25 predicted miRNA-related diseases were verified by literature and dbDEMC database. It is anticipated that Seq-SymRF could be regarded as a powerful high-throughput virtual screening tool for drug research and development. All source codes can be downloaded from https://github.com/LeeKamlong/Seq-SymRF.

Identification of inhibitors targeting HIF-2α/c-Myc by molecular docking and MM-GBSA technology

The treatment of ccRCC by targeting hypoxia-inducible factor HIF-2α is currently a direct and effective method. Studies have shown that HIF-2α and c-Myc cooperate to promote ccRCC tumor progression, and the overexpression of c-Myc is related to the progress and drug resistance of most human cancers. Although HIF-2α and c-Myc are important drug targets, their dual inhibitors are still lacking. We used virtual screening tools (mainly including molecular docking and MM-GBSA technology) to obtain some well-listed compounds that can potentially target HIF-2α and c-Myc and used molecular dynamics simulations to study their binding with these protein systems. Using a structure-based screening scheme, a batch of top-ranking compounds were selected, and their binding affinities were predicted of these compounds were performed. Representative compound C93106, C43257, and C41580 all showed good comprehensive binding score. Our results indicate that the target compounds can all form key interactions with the active site of the protein, and 30 ns molecular dynamic simulation of the complex system indicates a stable binding conformation. This research laid the foundation for the development of more effective and specific HIF-2α and c-Myc dual-target inhibitors.

Systemic In Silico Screening in Drug Discovery for Coronavirus Disease (COVID-19) with an Online Interactive Web Server.

The emergence of the new coronavirus (nCoV-19) has impacted human health on a global scale, while the interaction between the virus and the host is the foundation of the disease. The viral genome codes a cluster of proteins, each with a unique function in the event of host invasion or viral development. Under the current adverse situation, we employ virtual screening tools in searching for drugs and natural products which have been already deposited in DrugBank in an attempt to accelerate the drug discovery process. This study provides an initial evaluation of current drug candidates from various reports using our systemic in silico drug screening based on structures of viral proteins and human ACE2 receptor. Additionally, we have built an interactive online platform (https://shennongproject.ai/) for browsing these results with the visual display of a small molecule docked on its potential target protein, without installing any specialized structural software. With continuous maintenance and incorporation of data from laboratory work, it may serve not only as the assessment tool for the new drug discovery but also an educational web site for the public.

Virtual echocardiography screening tool to differentiate hemodynamic profiles in pulmonary hypertension.

This study validated a novel virtual echocardiography screening tool (VEST), which utilized routinely reported echocardiography parameters to predict hemodynamic profiles in pulmonary hypertension (PH) and identify PH due to pulmonary vascular disease (PHPVD). Direct echocardiography imaging review has been shown to predict hemodynamic profiles in PH; however, routine use often overemphasizes Doppler-estimated pulmonary artery systolic pressure (PASPDE), which lacks discriminatory power among hemodynamically varied PH subgroups. In patients with PH of varying subtypes at a tertiary referral center, reported echocardiographic findings needed for VEST, including left atrial size, E:e' and systolic interventricular septal flattening, were obtained. Receiver operating characteristic analyses assessed the predictive performance of VEST vs. PASPDE in identifying PHPVD, which was later confirmed by right heart catheterization. VEST demonstrated far superior discriminatory power than PASPDE in identifying PHPVD. A positive score was 80.0% sensitive and 75.6% specific for PHPVD with an area under the curve of 0.81. PASPDE exhibited poorer discriminatory power with an area under the curve of 0.56. VEST's strong discriminatory ability remained unchanged when validated in a second cohort from another tertiary center. We demonstrated that this novel VEST using three routine parameters that can be easily extracted from standard echocardiographic reports can successfully capture PH patients with a high likelihood of PHPVD. During the Covid-19 pandemic, when right heart catheterization and timely access to experts at accredited PH centers may have limited widespread availability, this may assist physicians to rapidly and remotely evaluate PH patients to ensure timely and appropriate care.

Developing novel computational prediction models for assessing chemical-induced neurotoxicity using naïve Bayes classifier technique

Development of reliable and efficient alternative in vivo methods for evaluation of the chemicals with potential neurotoxicity is an urgent need in the early stages of drug design. In this investigation, the computational prediction models for drug-induced neurotoxicity were developed by using the classical naïve Bayes classifier. Eight molecular properties closely relevant to neurotoxicity were selected. Then, 110 classification models were developed with using the eight important molecular descriptors and 10 types of fingerprints with 11 different maximum diameters. Among these 110 prediction models, the prediction model (NB-03) based on eight molecular descriptors combined with ECFP_10 fingerprints showed the best prediction performance, which gave 90.5% overall prediction accuracy for the training set and 82.1% concordance for the external test set. In addition, compared to naïve Bayes classifier, the recursive partitioning classifier displayed worse predictive performance for neurotoxicity. Therefore, the established NB-03 prediction model can be used as a reliable virtual screening tool to predict neurotoxicity in the early stages of drug design. Moreover, some structure alerts for characterizing neurotoxicity were identified in this research, which could give an important guidance for the chemists in structural modification and optimization to reduce the chemicals with potential neurotoxicity.