External Prediction Research Articles

Rapid authentication of variants of Gastrodia elata Blume using near-infrared spectroscopy combined with chemometric methods

The variety is one of the most important factors to generate difference of chemical compositions, which unavoidably influences the quality of natural medicine. Thus, simple and rapid authentication of different variants has great academic and practical significance. In this study, the goal was achieved with the help of near infrared spectroscopy (NIR) and chemometrics by using Gastrodia elata Blume as an example. A total of 540 samples including two classes of variants and their forms were investigated as a whole. The mean spectra of samples of each class and their 2-D synchronous correlation spectra were simultaneously applied to discover the difference of chemical characteristics. After hybrid pre-processing of the first and second derivative combined with Savitzky-Golay and Norris filtering, partial least squares discrimination analysis (PLS-DA) on the basis of latent variable projection was used to assess the feasibility for classification. The results show higher prediction accuracy in both internal test set and external prediction set. In order to further improve the robustness for modeling, three methods for wavelength selection were comprehensively compared to optimize PLS-DA models, including variable importance in the projection (VIP), random frog (RF), and Monte Carlo uninformative variable elimination (MC-UVE). The prediction accuracy of combination of the 2nd derivative, Norris, MC-UVE and PLS-DA achieved to 99.11% and 98.89% corresponding to the internal test set and external prediction set, respectively. The strategies proposed in this work perform effectiveness for rapid and accurate authentication of variants of plants with high chemical complexity.

An improved successive projections algorithm version to variable selection in multiple linear regression

The aim of the successive projections algorithm (SPA) is to enhance the accuracy of multiple linear regressions (MLR) by minimizing the impact of collinearity effects in the calibration data set. Combining SPA with MLR as a variable selection approach has resulted in the SPA-MLR method, which has been reported in literature to produce models with good prediction ability compared to conventional full-spectrum models obtained with partial-least-squares (PLS) in some cases. This paper proposes the addition of a filter step to the current version of the SPA algorithm to reduce the number of uninformative variables before the projection phase and assist the algorithm in selecting the best variables on subsequent steps. The proposed fSPA-MLR algorithm is evaluated in two case studies involving the near-infrared spectrometric analysis of pharmaceutical tablet and diesel/biodiesel mixture samples. Compared to PLS, the fSPA-MLR models demonstrate similar or better performance. Moreover, the fSPA-MLR models outperform the original SPA-MLR in both cross-validation and external prediction. The fSPA-MLR models deliver superior results regardless of the pre-processing algorithm tested, including first-derivative Savitzky-Golay (SG) and Standard Normal Variate (SNV), or even in raw spectra data.

In-silico design of some pyrazolopyridine analogs as breast cancer inhibitors: QSAR modeling, molecular docking, MD simulation, and pharmacological features profiling

In this work, the concepts of in silico approaches were utilized to design novel anti-breast cancer molecules using pyrazolopyridine derivatives. A quantitative structure-activity relationship (QSAR) model was developed using ATS0s, ATS7s, MATS6c, MDEC-33, and MDEN-22 descriptors. The model exhibits internal and external validation parameters such as a correlation coefficient (R2) value of 0.930, an adjusted correlation coefficient (R2adj) value of 0.907, a cross validation coefficient (Q2cv) value of 0.851, and an external prediction coefficient (R2ext) value of 0.716. These values passed the minimum recommended value for an acceptable QSAR model, confirming the model's robustness and predictability. With compound 26 (pIC50 = 5.62) as the template, ten compounds with activities ranging from pIC50 = 7.07 to 7.83 were designed. The results of molecular docking of the designed molecules with the estrogen receptor+ (ER+) active site residues showed better interactions with binding affinities ranging from −150.839 and −156.332 kcal/mol compared to the template (docking score = −151.206 kcal/mol) and Tamoxifen (docking score = −147.69 kcal/mol). The dynamic behavior of the ligand-protein complexes was studied via a 100-ns molecular dynamic simulation, and the results revealed that ligand binding was found to be strong enough so that the ligand doesn't leave its active site throughout the simulation with an RMSD value less than 0.25 Å. Additionally, the protein Cα RMSF showed that the active site residues have low fluctuation and a lower and constant Rg value of the ligand, further confirming the ligand-protein stability. The outcomes of the drug-likeness and ADMET studies demonstrated the pharmacological potency of the hypothetical molecules. Hence, the designed compounds could be utilized as novel anti-breast cancer drug candidates.

Predictive Quantitative Read-Across Structure-Property Relationship Modeling of the Retention Time (Log tR) of Pesticide Residues Present in Foods and Vegetables.

The retention time (log tR) of pesticidal compounds in a reverse-phase high-performance liquid chromatography (HPLC) analysis has a direct relationship with lipophilicity, which could be related to the ecotoxicity potential of the compounds. The novel quantitative read-across structure-property relationship (q-RASPR) modeling approach uses similarity-based descriptors for predictive model generation. These models have been shown to enhance external predictivity in previous studies for several end points. The current study describes the development of a q-RASPR model using experimental retention time data (log tR) in the HPLC experiments of 823 environmentally significant pesticide residues collected from a large compound database. To model the retention time (log tR) end point, 0D-2D descriptors have been used along with the read-across-derived similarity descriptors. The developed partial least squares (PLS) model was rigorously validated by various internal and external validation metrics as recommended by the Organization for Economic Co-operation and Development (OECD). The final q-RASPR model is proven to be a good fit, robust, and externally predictive (ntrain = 618, R2 = 0.82, Q2LOO = 0.81, ntest = 205, and Q2F1 = 0.84) that literally outperforms the external predictivity of the previously reported quantitative structure-property relationship (QSPR) model. From the insights of modeled descriptors, lipophilicity is found to be the most important chemical property, which positively correlates with the retention time (log tR). Various other characteristics, such as the number of multiple bonds (nBM), graph density (GD), etc., have a substantial and inversely proportionate relationship with the retention time end point. The software tools utilized in this study are user-friendly, and most of them are free, which makes our methodology quite cost-effective when compared to experimentation. In a nutshell, to obtain better external predictivity, interpretability, and transferability, q-RASPR is an efficient technique that has the potential to be employed as a good alternative approach for retention time prediction and ecotoxicity potential identification.

Neutralization evolution of concrete under acid rain and carbonation erosion: a review

Concrete structures in atmospheric environments are constantly affected by acid rain and carbon dioxide erosion, and the evolution of concrete neutralization is an important factor influencing the long-term performance of the material. To deepen our understanding of the long-term performance evolution mechanisms of concrete, this study reviews the mechanism of acid rain and carbonation neutralization, summarizes studies regarding the evolution of the macroscopic properties and microstructures of concrete, and discusses the internal and external influencing factors and prediction models for concrete neutralization evolution. A thorough analysis of existing studies reveals that the microscopic properties and microstructures of concrete undergo a process of optimization followed by decay in response to acid rain neutralization. The expansion products from acid rain can lead to the deterioration of concrete microstructure, whereas carbonation neutralization products enhance microstructural compactness. Moreover, the study found that the degree of concrete neutralization increases with an increase in the water-cement ratio (W/C). Furthermore, the maximal degree of concrete neutralization occurs within a relative humidity ranging from 45% to 65%. According to the diffusion theory, the process of concrete neutralization is an evolutionary one that operates from the outside to the inside of the material and is influenced by various factors. The aim of this review is to provide scholars with a systematic understanding of the evolution of concrete neutralization under acid rain and carbonation, and to establish a theoretical foundation for future research in this area. Potential areas for future research on concrete neutralization are also described in this review.

A novel adaptive dynamic GA combined with AM to optimize ANN for multi-output prediction: Small samples enhanced in industrial processing

Multi-output prediction based on small samples (SSs) produced in industrial processing plays an important role in improving processing efficiency and reducing production costs. However, insufficient sample capacity will result in inaccurate and unstable multi-output regression prediction. To solve this problem, this paper jointly optimizes an artificial neural network (ANN) by employing an adaptive dynamic genetic algorithm and adaptive moment estimation (ADGA-AM-ANN) based on enhanced SSs. The data are enhanced by adding noise to the output data to prevent overfitting due to insufficient sample capacity. The adaptive dynamic genetic algorithm is employed to grope for the optimal parameters of the ANN in the global range. This prevents the ANN from falling into a local minimum in the training process, so as to improve the overall accuracy of the model in simultaneously predicting multiple variables, and to promote its internal prediction stability. Adaptive moment estimation is employed to optimize the gradient of optimal parameters to accelerate convergence and deepen the external prediction stability of the model. The cases of cutting and laser processes are used to verify the accuracy and stability of the proposed method.

SA6 The Utility of “Single Exposure” Versus Causal Architecture Approaches in Real-World Research

Over recent decades, epidemiologic methodology has evolved substantially to encompass a gamut of techniques for focalizing on the most unbiased effect estimate of a single exposure (e.g., a treatment effect). By contrast, less attention has been paid to the refinement of study designs and analytic techniques intended to encompass the underlying milieu of co-occurring clinical, environmental, and social determinants that interactively influence disease risk or outcomes. Single exposure approaches (i.e., the “potential outcomes framework”) have been criticized as being reductionist and have long been debated in the broader epidemiologic community, despite the utility of this approach in optimizing internal validity. Further, the gap between clinical trial efficacy and real-world (RW) effectiveness remains a concern for both regulatory entities and payers, though this gap may be partly attributable to varying prevalences of important effect modifiers between trial and RW patient populations. This conceptual paper will expand upon the existing conceptual paradigms by applying them to RW research and delineating key considerations related to assessing whether a single exposure or causal architecture approach would be fit-for-purpose based on the overarching research question. Drawing upon examples from RW research types, such as use of external control groups, comparative effectiveness, and prediction of likely responders to treatment, the strengths and weaknesses of each approach will be demonstrated using case studies related to regulatory and/or payer decision-making. Methodologic recommendations will be provided. Ultimately, both single exposure and causal architecture approaches to study design and analysis have a place in pharmacoepidemiology and real-world research. A careful examination of when each approach is fit-for-purpose can lead to the application of innovative strategies to the design and conduct of RW studies.

2D & 3D-QSAR Studies on a Series of Quinoline-Amino-piperidine Derivatives as Potent Mycobacterium DNA-Gyrase-B Inhibitors

Introduction: Mycobacterium tuberculosis is a familiar infectious bacillus that causes tuberculosis, which primarily affects the lungs and the spinal cord. To combat the growing difficulties in treating MTB, it is necessary to create safe medications with novel mechanisms of action.
 Objective: To design and develop some novel quinolone-amino piperidine derivatives with potent mycobacterium DNAgyraseB inhibitory using the QSAR technique.
 Methods: Multiple linear regression (MLR), partial least squares (PLS), and k-nearest neighbour molecular field analysis ((kNN-MFA) were utilised in the development of 2D and 3D-QSAR models, respectively; these models were then validated. 
 Results: The recently developed 2D-QSAR model can explain 85.07% (r2 = 0.8507) of the total variance incorporated into the training set. In addition, the model has an internal prediction capacity (q2) of 77.65% and an external prediction capacity (pred r2) of 83.64%, respectively. The F test confirms that the likelihood of the model failing is extremely low. The 3D-QSAR model explains the values of k (2), q2 = 0.5707, pred r2 = 0.7843, q2 se = 0.3167, and pred r2 se = 0.3111. This demonstrates that the QSAR equation obtained in that way is statistically significant and that the model has a predictive capacity of 78.43%.
 Conclusion: The robustness of the developed 2D or 3D-QSAR models provides the necessary information and is expected to provide an excellent option for drug design.

Design of novel DABO derivatives as HIV-1 RT inhibitors using molecular docking, molecular dynamics simulations and ADMET properties

HIV-1 reverse transcriptase is an important target for developing effective anti-HIV-1 inhibitors. Different types of small molecules have been designed based on this target, showing different levels of inhibitory activity against various types of HIV-1 strains. The relationship between structure and activity of DABO derivatives was investigated by means of 3D-QSAR molecular model, molecular docking, molecular dynamics and ADMET properties. The statistical results of molecular models show that the CoMFA and CoMSIA models have good internal stability (CoMFA: q2 = 0.623, r2 = 0.946; CoMSIA: q2 = 0.668, r2 = 0.983) and external prediction ability (CoMFA: rpred 2 = 0.961; CoMSIA: rpred 2 = 0.961). In addition, molecular docking has explored the mechanism of action between small molecules and receptor proteins, and the results show that hydrogen bonding between amino acid Lys101 and small molecules can improve the affinity of ligands to receptor binding. A total of 12 novel molecules were designed and their activities were predicted based on the 3D-QSAR model and molecular docking results. The results showed that the designed molecules had higher predictive activity. Subsequently, 100 ns MD simulation and binding free energy verified the stability of molecular docking results. Finally, the pharmacokinetic properties of the novel designed molecule were verified by using ADMET to predict its properties. These results can provide reference for the design and development of novel and effective HIV-1 RT inhibitors, and provide new ideas for the design of subsequent drugs. Communicated by Ramaswamy H. Sarma

QSAR‐Guided Study for the Microwave‐Assisted Synthesis of 4‐Methylquinoline Derivatives with Antimycobacterial Activity

AbstractTuberculosis was discovered more than a century ago, and it is still a disease that presents difficulties in its treatment due to the appearance of new resistant strains. To design new antituberculosis agents with a quinoline structure, our group developed three‐dimensional structure‐activity relationship (3D‐QSAR) models, based on Comparative Molecular Field Analysis (CoMFA) and Comparative Molecular Similarity Index Analysis (CoMSIA). Statistically robust models (q2>0.6; r2ncv>0.8; r2pred>0.7) and with good external predictability were obtained. We found that positions 2, 3, and 4 of the quinoline nucleus are directly related to the modulation of the growth inhibitory activity of Mycobacterium tuberculosis H37Rv. To validate the models, we synthesized twelve quinolines through the Friëdlander reaction and three indolinones. The synthesized compounds were evaluated in the growth inhibition of resistant H37Rv tuberculosis strains (rpoBS450L, katGdel, and gyrAD94K) and in nontuberculous strains (M. avium and M. abscessus). We found that the compound (Z)‐4‐((2‐oxoindolin‐3‐ylidene)amino)‐N‐(thiazol‐2‐yl)benzenesulfonamide was active in all resistant strains; the compound 2‐(1H‐indol‐3‐yl)‐4‐methylquinoline was active on M. avium; and the compound 10‐methyl‐11H‐indeno[1,2‐b]quinoline was active on M. abscessus. These models are useful for the discovery of new compounds with inhibitory properties of M. tuberculosis H37Rv, and with potential applications in resistant and nontuberculous strains.

Numerical analysis of unsteady flow features around a realistic Estate vehicle with hybrid RANS/LES methods

A comparative study is conducted to analyze the flow around a realistic Peugeot Estate car (308 SW) using RKE and SST RANS models, as well as DDES hybrid RANS/LES approach. The authors used previous work on 25° Ahmed bodies to establish the best grid and DDES parameters for external aerodynamics prediction. The aim of this work is to extend and transpose the numerical methodology to a realistic car. Numerical and physical characteristics related to the simulation of this complex flow are therefore carefully investigated. The numerical results are compared to in-house experiments carried out in S2A Stellantis Wind Tunnel. Two grids with 115 and 185 million elements, based on 20 prism layers cells near the wall, are used to compare the three turbulence models. Even with a smaller grid size in the wake and underbody, the predictions of drag and lift coefficients are almost identical. Moreover, refining the near-wall region from 20 to 30 prism layers has no effect on drag coefficient and flow prediction. Consequently, the grid with 20 prism layers and 115M cells is used as reference grid in this work. Then, the flow features between RKE, SST and SST DDES models are compared to two-dimensional (2D) and three-dimensional (3D) experimental results. RANS gives better agreement on aerodynamics forces. However, DDES reproduces accurately the specific three-dimensional topology of the flow around the car, in the wake and the underbody when RANS results are globally wrong.

Predicting the ecotoxicity of endocrine disruptive chemicals: Multitasking in silico approaches towards global models

Manufactured substances known as endocrine disrupting chemicals (EDCs) released in the environment, through the use of cosmetic products or pesticides, can cause severe eco and cytotoxicity that may induce trans-generational as well as long-term deleterious effects on several biological species at relatively low doses, unlike other classical toxins. As the need for effective, affordable and fast EDCs environmental risk assessment has become increasingly pressing, the present work introduces the first moving average-based multitasking quantitative structure-toxicity relationship (MA-mtk QSTR) modeling specifically developed for predicting the ecotoxicity of EDCs against 170 biological species belonging to six groups. Based on 2,301 data-points with high structural and experimental diversity, as well as on the usage of various advanced machine learning methods, the novel most predictive QSTR models display overall accuracies > 87% in both training and prediction sets. However, maximum external predictivity was achieved when a new multitasking consensus modeling approach was applied to these models. Additionally, the developed linear model provided means to investigate the determining factors for eliciting higher ecotoxicity by the EDCs towards different biological species, identifying several factors such as solvation, molecular mass and surface area as well as the number of specific molecular fragments (e.g.: aromatic hydroxy and aliphatic aldehyde). The resource to non-commercial open-access tools to develop the models is a useful step towards library screening to speed up regulatory decision on discovery of safe alternatives to reduce the hazards of EDCs.

Machine Learning‐Enabled Virtual Screening with Multiple Protein Structures toward the Discovery of Novel JAK3 Inhibitors: Integration of Molecular Docking, Pharmacophore, and Naïve Bayesian Classification

AbstractExtensive research has accumulated suggesting that Janus kinase 3 (JAK3) is closely related to the occurrence and development of various human diseases, making JAK3 a highly potential drug target. However, JAK3 has high homology with other members of the JAK family, making the development of JAK3 inhibitors full of challenges. Thus, here, a naïve Bayesian classification (NBC) model based on multiple JAK3 protein conformations, which integrates molecular docking, pharmacophore, and molecular descriptors, is developed to find novel JAK3 inhibitors. First, the validation set is used to prove whether molecular docking or pharmacophore, integrating multiple JAK3 conformations always has higher prediction accuracy than that of any single conformation. Second, external prediction reveals that the NBC model combining molecular docking, pharmacophore, and important molecular features could significantly improve the enrichment of active JAK3 inhibitors. Finally, the optimal NBC model is utilized for virtual screening against a large chemical database and some compounds with high Bayesian scores are identified. Altogether, the machine learning‐based virtual screening protocol not only has strong efficiency but also has high screening accuracy. It is hoped that the developed virtual screening strategy could provide valuable guidance for the discovery of novel JAK3 inhibitors.

Machine-learning-based similarity meets traditional QSAR: “q-RASAR” for the enhancement of the external predictivity and detection of prediction confidence outliers in an hERG toxicity dataset

Recently, the concept of quantitative Read-Across Structure-Activity Relationship (q-RASAR) has been introduced by using various Machine Learning (ML) - derived similarity functions in the traditional quantitative structure-activity relationship (QSAR) modeling framework with the objective of enhancing the external predictivity of models while using the same available chemical information content. The present study uses the hERG K+ channel inhibition cardiotoxicity, a pharmaceutically relevant endpoint, as the modeling set for making predictions using the novel q-RASAR approach, as the approach combines the merits of QSAR and Read-Across, and generates simple and interpretable models using various similarity and error-based measures as descriptors. The cardiotoxicity data (in terms of pIC50 values) were collected from the literature. The curated data set was then divided into training and test sets using the sorted response-based division algorithm. The important set of features was identified based on the internal validation metrics of initial genetic algorithm models. Based on the features selected in the final Multiple Linear Regression (MLR) model, RASAR descriptors were computed using a tool available from https://sites.google.com/jadavpuruniversity.in/dtc-lab-software/home. The RASAR descriptors were then merged with the previously selected features, and an MLR q-RASAR model was generated using the grid search approach. The prediction outliers were then identified using the novel DTC Applicability Domain Plot, and the q-RASAR models were used for predictions after the removal of prediction outliers. A final Partial Least Squares (PLS) q-RASAR model was generated to obviate inter-correlation among descriptors. Various other Machine Learning approaches were also employed with the optimization of relevant hyperparameters based on the cross-validation approach, and the final test set prediction results were compared. Based on the performance in the test set predictions and interpretability, the PLS q-RASAR model was chosen as the final model which provided enhanced predictivity in comparison to previously reported models even without using 3-D descriptors. This model can thus be used for the quick screening of molecules, even before their synthesis, to estimate their cardiotoxic potential thus prioritizing molecules for further experimental testing in the drug discovery pipeline. A Java-based prediction tool has also been developed for the quick screening of cardiotoxic properties of query compounds and made available from https://sites.google.com/jadavpuruniversity.in/dtc-lab-software/home.

Ecotoxicological QSAR study of fused/non-fused polycyclic aromatic hydrocarbons (FNFPAHs): Assessment and priority ranking of the acute toxicity to Pimephales promelas by QSAR and consensus modeling methods

Fused/non-fused polycyclic aromatic hydrocarbons (FNFPAHs) have a variety of toxic effects on ecosystems and human body, but the acquisition of their toxicity data is greatly limited by the limited resources available. Here, we followed the EU REACH regulation and used Pimephales promelas as a model organism to investigate the quantitative structure-activity relationship (QSAR) between the FNFPAHs and their toxicity for the aquatic environment for the first time. We developed a single QSAR model (SM1) containing five simple and interpretable 2D molecular descriptors, which met the validation of OECD QSAR-related principles, and analyzed their mechanistic relationships with toxicity in detail. The model had good degree of fitting and robustness, and had better external prediction performance (MAEtest = 0.4219) than ECOSAR model (MAEtest = 0.5614). To further enhance its prediction accuracy, the three qualified single models (SMs) were used for constructing consensus models (CMs), the best one CM2 (MAEtest = 0.3954) had a significantly higher prediction accuracy for test compounds than SM1, and also outperformed the T.E.S.T. consensus model (MAEtest = 0.4233). Subsequently, the toxicity of 252 true external FNFPAHs from Pesticide Properties Database (PPDB) was predicted by SM1, the prediction results showed that 94.84 % compounds were reliably predicted within the model's application domain (AD). We also applied the best CM2 to predict the untested 252 FNFPAHs. Furthermore, we provided a mechanistic analysis and explanation for pesticides ranked as top 10 most toxic FNFPAHs. In summary, all developed QSAR and consensus models can be used as efficient tools for predicting the acute toxicity of unknown FNFPAHs to Pimephales promelas, thus being important for the risk assessment and regulation of FNFPAHs contamination in aquatic environment.

On Some Novel Similarity-Based Functions Used in the ML-Based q-RASAR Approach for Efficient Quantitative Predictions of Selected Toxicity End Points.

The novel quantitative read-across structure-activity relationship (q-RASAR) approach uses read-across-derived similarity functions in the quantitative structure-activity relationship (QSAR) modeling framework in a unique way for supervised model generation. The aim of this study is to explore how this workflow enhances the external (test set) prediction quality of conventional QSAR models by the incorporation of some novel similarity-based functions as additional descriptors using the same level of chemical information. To establish this, five different toxicity data sets, for which QSAR models were reported previously, have been considered in the q-RASAR modeling exercise, which uses chemical similarity-derived measures. The identical sets of chemical features along with the same compositions of training and test sets as reported previously were used in the present analysis for ease of comparison. The RASAR descriptors were calculated based on a chosen similarity measure with the default setting of relevant hyperparameter(s) and were then clubbed with the original structural and physicochemical descriptors, and the number of selected features was further optimized by employing a grid search technique applied on the respective training sets. These features were then used to develop multiple linear regression (MLR) q-RASAR models that show enhanced predictivity as compared to the QSAR models developed previously. Moreover, various other ML algorithms like support vector machine (SVM), linear SVM, random forest, partial least squares, and ridge regression were also employed using the same feature combinations as used in the MLR models to compare the prediction qualities. The q-RASAR models for five different data sets possess at least one of the RASAR descriptors, RA function, gm, and average similarity, suggesting that these are important determinants of similarities that contribute to the development of predictive q-RASAR models, as also evident from the SHAP analysis of the models.

A machine learning q-RASPR approach for efficient predictions of the specific surface area of perovskites.

In this study, the specific surface area of various perovskites was modeled using a novel quantitative read-across structure-property relationship (q-RASPR) approach, which clubs both Read-Across (RA) and quantitative structure-property relationship (QSPR) together. After optimization of the hyper-parameters, certain similarity-based error measures for each query compound were obtained. Clubbing some of these error-based measures with the previously selected features along with the Read-Across prediction function, a number of machine learning models were developed using Partial Least Squares (PLS), Ridge Regression (RR), Linear Support Vector Regression (LSVR), Random Forest (RF) regression, Gradient Boost (GBoost), Adaptive Boosting (Adaboost), Multiple Layer Perceptron (MLP) regression and k-Nearest Neighbor (kNN) regression. Based on the repeated cross-validation as well as external prediction quality and interpretability, the PLS model (nTraining = 38, nTest = 12, =0.737, was selected as the best predictor which underscored the previously reported results. The finally selected model should efficiently predict specific surface areas of other perovskites for their use in photocatalysis. The new q-RASPR method also appears promising for the prediction of several other property endpoints of interest in materials science.

Efficient predictions of cytotoxicity of TiO2-based multi-component nanoparticles using a machine learning-based q-RASAR approach

The availability of experimental nanotoxicity data is in general limited which warrants both the use of in silico methods for data gap filling and exploring novel methods for effective modeling. Read-Across Structure-Activity Relationship (RASAR) is an emerging cheminformatic approach that combines the usefulness of a QSAR model and similarity-based Read-Across predictions. In this work, we have generated simple, interpretable, and transferable quantitative-RASAR (q-RASAR) models which can efficiently predict the cytotoxicity of TiO2-based multi-component nanoparticles. A data set of 29 TiO2-based nanoparticles with specific amounts of noble metal precursors was rationally divided into training and test sets, and the Read-Across-based predictions for the test set were generated. The optimized hyperparameters and the similarity approach, which yield the best predictions, were used to calculate the similarity and error-based RASAR descriptors. A data fusion of the RASAR descriptors with the chemical descriptors was done followed by the best subset feature selection. The final set of selected descriptors was used to develop the q-RASAR models, which were validated using the stringent OECD criteria. Finally, a random forest model was also developed with the selected descriptors, which could efficiently predict the cytotoxicity of TiO2-based multi-component nanoparticles superseding previously reported models in the prediction quality thus showing the merits of the q-RASAR approach. To further evaluate the usefulness of the approach, we have applied the q-RASAR approach also to a second cytotoxicity data set of 34 heterogeneous TiO2-based nanoparticles which further confirmed the enhancement of external prediction quality of QSAR models after incorporation of RASAR descriptors.

Evaluation of Model for External Short Circuit Voltage Behavior Prediction of Lithium-ion Batteries

Evaluation of Model for External Short Circuit Voltage Behavior Prediction of Lithium-ion Batteries

Validation of external cephalic version prediction model in population with low neuraxial anesthesia use

Validation of external cephalic version prediction model in population with low neuraxial anesthesia use