Double Cross-validation Research Articles

Evaluation of grade and invasiveness of bladder urothelial carcinoma using infrared imaging and machine learning.

Urothelial bladder carcinoma (BC) is primarily diagnosed with a subjective examination of biopsies by histopathologists, but accurate diagnosis remains time-consuming and of low diagnostic accuracy, especially for low grade non-invasive BC. We propose a novel approach for high-throughput BC evaluation by combining infrared (IR) microscopy of bladder sections with machine learning (partial least squares-discriminant analysis) to provide an automated prediction of the presence of cancer, invasiveness and grade. Cystoscopic biopsies from 50 patients with clinical suspicion of BC were histologically examined to assign grades and stages. Adjacent tissue cross-sections were IR imaged to provide hyperspectral datasets and cluster analysis segregated IR images to extract the average spectra of epithelial and subepithelial tissues. Discriminant models, which were validated using repeated random sampling double cross-validation, showed sensitivities (AUROC) ca. 85% (0.85) for the identification of cancer in epithelium and subepithelium. The diagnosis of non-invasive and invasive cases showed sensitivity values around 80% (0.84-0.85) and 76% (0.73-0.80), respectively, while the identification of low and high grade BC showed higher sensitivity values 87-88% (0.91-0.92). Finally, models for the discrimination between cancers with different invasiveness and grades showed more modest AUROC values (0.67-0.72). This proves the high potential of IR imaging in the development of ancillary platforms to screen bladder biopsies.

Abstract PR002: Genomic predictor can discriminate between high- and low-risk DCIS

Abstract Introduction: Ductal carcinoma in situ (DCIS) is considered a non-obligate precursor of invasive ductal carcinoma. With the aim of preventing a subsequent invasive cancer, all DCIS lesions are currently treated with surgical excision often supplemented with radiotherapy (RT). To prevent DCIS over- or undertreatment, a reliable marker of DCIS invasiveness risk is urgently needed. Methods: We studied two large DCIS cohorts: the Sloane cohort, a prospective breast screening cohort from the UK (median follow-up of 11 years), and a Dutch population-based cohort (NKI, median follow-up of 13 years). FFPE tissue specimens from patients with pure primary DCIS after breast-conserving surgery (BCS) +/- RT that did develop a subsequent ipsilateral event (DCIS or invasive) were considered as cases, whereas patients that did not develop any form of recurrence up to the last follow-up or death were considered as controls. We performed copy number analysis (CNA) and RNAseq analysis on 229 cases (80 DCIS only recurrences) and 344 controls. Results: DCIS was classified into the PAM50 subtypes using RNAseq data which revealed an enrichment of luminal A phenotype in DCIS that did not recur (P = 0.01, Fisher Exact test). No single copy number aberration was more common in cases compared to controls. RNAseq data did not reveal any genes significantly over/under-expressed in cases versus controls after FDR correction. However, by limiting the analysis to samples that had not had RT and excluding pure DCIS recurrences, we could develop a penalized Cox model from RNAseq data. The model was trained on weighted samples (to correct for the biased sampling of the case-control dataset) from the NKI series with double loop cross-validation. The genes were selected using the Elastic net framework of penalization. Using this predicted hazard ratio, the samples were split into high, medium, and low-risk quantiles, with a recurrence risk of 23%, 7% and 2%, respectively at 5 years (p = 10-10, Wald test). The NKI-trained predictor was independently validated in the Sloane No RT no DCIS recurrence cohort (p = 0.02, Wald test). GSEA analysis revealed proliferation hallmarks enriched in the recurrence predictor (FDR = 0.058). The RNAseq predictor was more predictive of recurrence than PAM50, clinical features (Grade, Her2 and ER) and the 12-gene Oncotype DCIS score (p < 0.001, permutation test using the Wald statistic) in both the NKI and Sloane series. Conclusion: Genomic profiling of two independent series of DCIS with outcome data did not reveal any clear associations with recurrence until analysis was limited to a set of samples who had not had radiotherapy and DCIS recurrences were excluded. We then identified an RNAseq-based classifier that could differentiate primary DCIS in low-, medium-, and high-risk groups, and validated it in an independent cohort. This classifier, if validated in other datasets, will allow us to identify women who do not need intensive treatment for their DCIS. Citation Format: Maria Roman Escorza, Michael Sheinman, Tycho Bismeijer, Ahmed A. Ahmed, Vandna Shah, Jeffrey R. Marks, Lorraine M. King, Anargyros Megalios, Lindy L. Visser, Marlous Hoogstrat, Helen R. Davies, Tapsi Kumar, Deborah Collyar, Hilary Stobart, Sarah Pinder, Nicholas N. Navin, Andrew Futreal, Serena Nik-Zainal, E. Shelley Hwang, Esther H. Lips, Alastair Thompson, Lodewyk F.A. Wessels, Jelle Wesseling, Elinor J. Sawyer. Genomic predictor can discriminate between high- and low-risk DCIS [abstract]. In: Proceedings of the AACR Special Conference on Rethinking DCIS: An Opportunity for Prevention?; 2022 Sep 8-11; Philadelphia, PA. Philadelphia (PA): AACR; Can Prev Res 2022;15(12 Suppl_1): Abstract nr PR002.

Development of QSRR model for hydroxamic acids using PCA-GA-BP algorithm incorporated with molecular interaction-based features.

As a potent zinc chelator, hydroxamic acid has been applied in the design of inhibitors of zinc metalloenzyme, such as histone deacetylases (HDACs). A series of hydroxamic acids with HDAC inhibitory activities were subjected to the QSRR (Quantitative Structure-Retention Relationships) study. Experimental data in combination with calculated molecular descriptors were used for the development of the QSRR model. Specially, we employed PCA (principal component analysis) to accomplish dimension reduction of descriptors and utilized the principal components of compounds (16 training compounds, 4 validation compounds and 7 test compounds) to execute GA (genetic algorithm)-BP (error backpropagation) algorithm. We performed double cross-validation approach for obtaining a more convincing model. Moreover, we introduced molecular interaction-based features (molecular docking scores) as a new type of molecular descriptor to represent the interactions between analytes and the mobile phase. Our results indicated that the incorporation of molecular interaction-based features significantly improved the accuracy of the QSRR model, (R2 value is 0.842, RMSEP value is 0.440, and MAE value is 0.573). Our study not only developed QSRR model for the prediction of the retention time of hydroxamic acid in HPLC but also proved the feasibility of using molecular interaction-based features as molecular descriptors.

Exploring quantitative structure-property relationship models for environmental fate assessment of petroleum hydrocarbons.

The rate and extent of biodegradation of petroleum hydrocarbons in the different aquatic environments is an important element to address. The major avenue for removing petroleum hydrocarbons from the environment is thought to be biodegradation. The present study involves the development of predictive quantitative structure-property relationship (QSPR) models for the primary biodegradation half-life of petroleum hydrocarbons that may be used to forecast the biodegradation half-life of untested petroleum hydrocarbons within the established models' applicability domain. These models use easily computable two-dimensional (2D) descriptors to investigate important structural characteristics needed for the biodegradation of petroleum hydrocarbons in freshwater (dataset 1), temperate seawater (dataset 2), and arctic seawater (dataset 3). All the developed models follow OECD guidelines. We have used double cross-validation, best subset selection, and partial least squares tools for model development. In addition, the small dataset modeler tool has been successfully used for the dataset with very few compounds (dataset 3 with 17 compounds), where dataset division was not possible. The resultant models are robust, predictive, and mechanisticallyinterpretable based on both internal and external validation metrics (R2 range of 0.605-0.959. Q2(Loo) range of 0.509-0.904, and Q2F1 range of 0.526-0.959). The intelligent consensus predictor tool has been used for theimprovement of the prediction quality for test set compounds which provided superior outcomes to those from individual partial least squares models based on several metrics (Q2F1 = 0.808 and Q2F2 = 0.805 for dataset 1 in freshwater). Molecular size and hydrophilic factor for freshwater, frequency of two carbon atoms at topological distance 4 for temperate seawater, and electronegative atom count relative to size for arctic seawater were found to be the most significant descriptors responsible for the regulation of biodegradation half-life of petroleum hydrocarbons.

Develop a diagnostic tool for dementia using machine learning and non-imaging features.

BackgroundEarly identification of Alzheimer’s disease or mild cognitive impairment can help guide direct prevention and supportive treatments, improve outcomes, and reduce medical costs. Existing advanced diagnostic tools are mostly based on neuroimaging and suffer from certain problems in cost, reliability, repeatability, accessibility, ease of use, and clinical integration. To address these problems, we developed, evaluated, and implemented an early diagnostic tool using machine learning and non-imaging factors.Methods and resultsA total of 654 participants aged 65 or older from the Nursing Home in Hangzhou, China were identified. Information collected from these patients includes dementia status and 70 demographic, cognitive, socioeconomic, and clinical features. Logistic regression, support vector machine (SVM), neural network, random forest, extreme gradient boosting (XGBoost), least absolute shrinkage and selection operator (LASSO), and best subset models were trained, tuned, and internally validated using a novel double cross validation algorithm and multiple evaluation metrics. The trained models were also compared and externally validated using a separate dataset with 1,100 participants from four communities in Zhejiang Province, China. The model with the best performance was then identified and implemented online with a friendly user interface. For the nursing dataset, the top three models are the neural network (AUROC = 0.9435), XGBoost (AUROC = 0.9398), and SVM with the polynomial kernel (AUROC = 0.9213). With the community dataset, the best three models are the random forest (AUROC = 0.9259), SVM with linear kernel (AUROC = 0.9282), and SVM with polynomial kernel (AUROC = 0.9213). The F1 scores and area under the precision-recall curve showed that the SVMs, neural network, and random forest were robust on the unbalanced community dataset. Overall the SVM with the polynomial kernel was found to be the best model. The LASSO and best subset models identified 17 features most relevant to dementia prediction, mostly from cognitive test results and socioeconomic characteristics.ConclusionOur non-imaging-based diagnostic tool can effectively predict dementia outcomes. The tool can be conveniently incorporated into clinical practice. Its online implementation allows zero barriers to its use, which enhances the disease’s diagnosis, improves the quality of care, and reduces costs.

QSAR, Molecular Docking, MD Simulation and MMGBSA Calculations Approaches to Recognize Concealed Pharmacophoric Features Requisite for the Optimization of ALK Tyrosine Kinase Inhibitors as Anticancer Leads.

ALK tyrosine kinase ALK TK is an important target in the development of anticancer drugs. In the present work, we have performed a QSAR analysis on a dataset of 224 molecules in order to quickly predict anticancer activity on query compounds. Double cross validation assigns an upward plunge to the genetic algorithm–multi linear regression (GA-MLR) based on robust univariate and multivariate QSAR models with high statistical performance reflected in various parameters like, fitting parameters; R2 = 0.69–0.87, F = 403.46–292.11, etc., internal validation parameters; Q2LOO = 0.69–0.86, Q2LMO = 0.69–0.86, CCCcv = 0.82–0.93, etc., or external validation parameters Q2F1 = 0.64–0.82, Q2F2 = 0.63–0.82, Q2F3 = 0.65–0.81, R2ext = 0.65–0.83 including RMSEtr < RMSEcv. The present QSAR evaluation successfully identified certain distinct structural features responsible for ALK TK inhibitory potency, such as planar Nitrogen within four bonds from the Nitrogen atom, Fluorine atom within five bonds beside the non-ring Oxygen atom, lipophilic atoms within two bonds from the ring Carbon atoms. Molecular docking, MD simulation, and MMGBSA computation results are in consensus with and complementary to the QSAR evaluations. As a result, the current study assists medicinal chemists in prioritizing compounds for experimental detection of anticancer activity, as well as their optimization towards more potent ALK tyrosine kinase inhibitor.

QSAR Evaluations to Unravel the Structural Features in Lysine-Specific Histone Demethylase 1A Inhibitors for Novel Anticancer Lead Development Supported by Molecular Docking, MD Simulation and MMGBSA.

Using 84 structurally diverse and experimentally validated LSD1/KDM1A inhibitors, quantitative structure–activity relationship (QSAR) models were built by OECD requirements. In the QSAR analysis, certainly significant and understated pharmacophoric features were identified as critical for LSD1 inhibition, such as a ring Carbon atom with exactly six bonds from a Nitrogen atom, partial charges of lipophilic atoms within eight bonds from a ring Sulphur atom, a non-ring Oxygen atom exactly nine bonds from the amide Nitrogen, etc. The genetic algorithm–multi-linear regression (GA-MLR) and double cross-validation criteria were used to create robust QSAR models with high predictability. In this study, two QSAR models were developed, with fitting parameters like R2 = 0.83–0.81, F = 61.22–67.96, internal validation parameters such as Q2LOO = 0.79–0.77, Q2LMO = 0.78–0.76, CCCcv = 0.89–0.88, and external validation parameters such as, R2ext = 0.82 and CCCex = 0.90. In terms of mechanistic interpretation and statistical analysis, both QSAR models are well-balanced. Furthermore, utilizing the pharmacophoric features revealed by QSAR modelling, molecular docking experiments corroborated with the most active compound’s binding to the LSD1 receptor. The docking results are then refined using Molecular dynamic simulation and MMGBSA analysis. As a consequence, the findings of the study can be used to produce LSD1/KDM1A inhibitors as anticancer leads.

Novel Assessment and Classification of Monthly Average Daily Global Solar Radiation Models Through a Figure of Merit Called Irradiation Time Equivalence: Analysis of 70 Regression Models Based on Air Temperature and Sunshine Hours Predictors

Abstract Irradiation time equivalence pioneers the classification of models that predict monthly average daily global solar radiation on a horizontal surface based on their double cross-validation performances. By exploiting indigenous irradiation data, novel irradiation-based models can be created and used to classify prediction models, thereby facilitating a deeper understanding of model performance beyond routine summary statistics. The concept was demonstrated by formulating novel 1-hour and 2-hour irradiation-based models to predict monthly average daily global horizontal irradiation. Double cross-validations of the two irradiation-based models and 70 existing regression models were performed using a pair of 5-year subsets. The 70 models used the measured meteorological predictors of air temperature and sunshine hours, either alone or combined. The irradiation time equivalence of a model evaluated under double cross-validation has been defined as the minimum number of hours of measured irradiation needed to predict the monthly average daily irradiation in an average year, with a root mean square error less than or equal to that of the model. Despite their intracompetitiveness, all 44 temperature-based models had an irradiation time equivalence of 1 h, while the remaining 26 models that contained the sunshine-hours predictor were classified with a higher-performance rank of 2 h. An irradiation time equivalence scale extending to 13 h was also developed to cater to the classification of higher-performance models. This fresh perspective on model performance directs future investigations toward determining whether prediction models exhibit global classification constancy.

Chemometric modeling of acute toxicity of diverse aromatic compounds against Rana japonica

Chemicals used in our daily life show different toxic effects to the aquatic and terrestrial species and thus hamper the ecological balance. In the present time, amphibians are one of them, which are threatened to be extinct. Quantitative structure-activity relationship (QSAR) is an useful tool for prediction involving less time, money and manpower without requiring any animal experiments to assess the unavailable acute toxicity data for the untested molecules. In this study, we have developed QSAR models for ecotoxicity of some waterborne diverse aromatic compounds on an amphibian species Rana japonica (Japanese brown frog) employing Genetic Algorithm (GA) for variable selection followed by Partial Least Squares (PLS) regression method following recommendations of the Organization for Economic Co-operation and Development (OECD) for QSAR model development. Double cross-validation (DCV) followed by Best Subset Selection (BSS) were employed to select suitable models. The models displayed promising statistical quality in terms of R2 (= 0.837–0.841), Q2LOO (= 0.782–0.787), R2pred or Q2F1 (= 0.802–0.82) and some other internal and external validation metrics for tadpoles of Rana japonica (NTraining = 44, NTest = 14). These models can be applied for data gap filling for a new untested compound falling within the applicability domain (AD) of the models.

Deep learning prediction of quantitative coronary angiography values using myocardial perfusion images with a CZT camera

PurposeEvaluate the prediction of quantitative coronary angiography (QCA) values from MPI, by means of deep learning. Methods546 patients (67% men) undergoing stress 99mTc-tetrofosmin MPI in a CZT camera in the upright and supine position were included (1092 MPIs). Patients were divided into two groups: ICA group included 271 patients who performed an ICA within 6 months of MPI and a control group with 275 patients with low pre-test probability for CAD and a normal MPI. QCA analyses were performed using radiologic software and verified by an expert reader. Left ventricular myocardium was segmented using clinical nuclear cardiology software and verified by an expert reader. A deep learning model was trained using a double cross-validation scheme such that all data could be used as test data as well. ResultsArea under the receiver-operating characteristic curve for the prediction of QCA, with > 50% narrowing of the artery, by deep learning for the external test cohort: per patient 85% [95% confidence interval (CI) 84%-87%] and per vessel; LAD 74% (CI 72%-76%), RCA 85% (CI 83%-86%), LCx 81% (CI 78%-84%), and average 80% (CI 77%-83%). ConclusionDeep learning can predict the presence of different QCA percentages of coronary artery stenosis from MPIs.

Artificial intelligence algorithms for the recognition of Brugada type 1 pattern on standard 12-leads ECG

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): This research project is funded by Tuscany Region Background/Introduction Electrocardiograms (ECGs) are rapidly moving from analog to digital versions. Consequently, a series of automatic analyses of standard 12-lead ECGs are attracting interest for their ability to support clinicians in the automatic recognition of specific features associated with different cardiac diseases [2]. Artificial Intelligence applications and Machine Learning (ML) algorithms have gained much attention in the last years for their ability to figure out patterns from data independently, without being explicitly taught rules. Peculiar features define the ECGs of patients with Brugada Syndrome (BrS); however, ambiguities still exist for the correct diagnosis of BrS and discrimination with respect to other pathologies. Purpose The BrAID (Brugada syndrome and Artificial Intelligence applications to Diagnosis) project aims to develop an innovative system for diagnosing Type 1 BrS based on ECG pattern recognition through the application of ML algorithms. In this work, an application of Echo State Networks (ESN), a type of Recurrent Neural Network (RNN), for the diagnosis of BrS from ECG is presented. Methods After approval from the Local Ethical Committees, 12-lead ECGs were obtained in patients enrolled in 5 Centers diagnosed with typical spontaneous Type 1 pattern (coved) (group A, 81 patients). Baseline ECG was also collected in patients undergoing the ajmaline test, classified as positive (group B, 37 patients) or negative (group C, 14 patients) according to test results. 174 patients with no clinical and familial history of arrhythmias were considered controls (group D). Data were collected from 4 beats extracted from the ECGs as input to the ESN. The datasets obtained in the different groups were used for the ESN model’s training and assessment (testing) through a double cross-validation approach. Results As shown in Table 1, the performances using three leads (V1, V2, V3) or V2 only were compared. The algorithm performance was assessed in all the datasets (group A+B+C+D) and in spontaneous BrS (group A) and controls (group D). A good accuracy (79.21%) was seen when the three leads were considered for groups A and D only; the best test set accuracy (80.20%) was obtained in the case in which V2 only was used as input in all the datasets. Conclusion(s) In this work, a novel system for diagnosing Type 1 BrS using an ESN approach was developed. Our preliminary results show that this ML model is able to detect ECG patterns associated with Type 1 BrS with good and comparable accuracy both when three leads (79.21% ) or V2 only (80.20%) were analyzed. The future availability of larger datasets could improve the model performance, increasing the ESN potentialities as a clinical support system tool to be used in everyday clinical practice. Table 1. The accuracy, specificity, and sensitivity reported for each dataset group are obtained through double cross-validation.

Linear and Non-Linear Soft Sensors for Predicting the Research Octane Number (RON) through Integrated Synchronization, Resolution Selection and Modelling.

The Research Octane Number (RON) is a key quality parameter for gasoline, obtained offline through complex, time-consuming, and expensive standard methods. Measurements are usually only available a few times per week and after long delays, making process control very challenging. Therefore, alternative methods have been proposed to predict RON from readily available data. In this work, we report the development of inferential models for predicting RON from process data collected in a real catalytic reforming process. Data resolution and synchronization were explicitly considered during the modelling stage, where 20 predictive linear and non-linear machine learning models were assessed and compared using a robust Monte Carlo double cross-validation approach. The workflow also handles outliers, missing data, multirate and multiresolution observations, and processes dynamics, among other features. Low RMSE were obtained under testing conditions (close to 0.5), with the best methods belonging to the class of penalized regression methods and partial least squares. The developed models allow for improved management of the operational conditions necessary to achieve the target RON, including a more effective use of the heating utilities, which improves process efficiency while reducing costs and emissions.

Comprehensive biomarker profiles and chemometric filtering of urinary metabolomics for effective discrimination of prostate carcinoma from benign hyperplasia

Prostate cancer (PCa) is the most commonly diagnosed cancer in male individuals, principally affecting men over 50 years old, and is the leading cause of cancer-related deaths. Actually, the measurement of prostate-specific antigen level in blood is affected by limited sensitivity and specificity and cannot discriminate PCa from benign prostatic hyperplasia patients (BPH). In the present paper, 20 urine samples from BPH patients and 20 from PCa patients were investigated to develop a metabolomics strategy useful to distinguish malignancy from benign hyperplasia. A UHPLC-HRMS untargeted approach was carried out to generate two large sets of candidate biomarkers. After mass spectrometric analysis, an innovative chemometric data treatment was employed involving PLS-DA classification with repeated double cross-validation and permutation test to provide a rigorously validated PLS-DA model. Simultaneously, this chemometric approach filtered out the most effective biomarkers and optimized their relative weights to yield the highest classification efficiency. An unprecedented portfolio of prostate carcinoma biomarkers was tentatively identified including 22 and 47 alleged candidates from positive and negative ion electrospray (ESI+ and ESI−) datasets. The PLS-DA model based on the 22 ESI+ biomarkers provided a sensitivity of 95 ± 1% and a specificity of 83 ± 3%, while that from the 47 ESI− biomarkers yielded an 88 ± 3% sensitivity and a 91 ± 2% specificity. Many alleged biomarkers were annotated, belonging to the classes of carnitine and glutamine metabolites, C21 steroids, amino acids, acetylcholine, carboxyethyl-hydroxychroman, and dihydro(iso)ferulic acid.

Prediction reliability of QSAR models: an overview of various validation tools.

The reliability of any quantitative structure-activity relationship (QSAR) model depends on multiple aspects such as the accuracy of the input dataset, selection of significant descriptors, the appropriate splitting process of the dataset, statistical tools used, and most notably on the measures of validation. Validation, the most crucial step in QSAR model development, confirms the reliability of the developed QSAR models and the acceptability of each step in the model development. The present review deals with various validation tools that involve multiple techniques that improve the model quality and robustness. The double cross-validation tool helps in building improved quality models using different combinations of the same training set in an inner cross-validation loop. This exhaustive method is also integrated for small datasets (< 40 compounds) in another tool, namely the small dataset modeler tool. The main aim of QSAR researchers is to improve prediction quality by lowering the prediction errors for the query compounds. 'Intelligent' selection of multiple models and consensus predictions integrated in the intelligent consensus predictor tool were found to be more externally predictive than individual models. Furthermore, another tool called Prediction Reliability Indicator was explained to understand the quality of predictions for a true external set. This tool uses a composite scoring technique to identify query compounds as 'good' or 'moderate' or 'bad' predictions. We have also discussed a quantitative read-across tool which predicts a chemical response based on the similarity with structural analogues. The discussed tools are freely available from https://dtclab.webs.com/software-tools or http://teqip.jdvu.ac.in/QSAR_Tools/DTCLab/ and https://sites.google.com/jadavpuruniversity.in/dtc-lab-software/home (for read-across).

Nitroaromatics as hypoxic cell radiosensitizers: A 2D-QSAR approach to explore structural features contributing to radiosensitization effectiveness

Hypoxia is the prime component of tumor microenvironment that plays a pivotal role in cancer progression. Nitroaromatic compounds are known to enhance the sensitivity of hypoxic cells to ionizing radiation. The application of computational tools like Quantitative Structure-Activity Relationship (QSAR) can be used to predict newly developed nitroaromatics or compounds with missing data. In the present work, three datasets consisting of 18 nitrofurans, 11 nitrothiophenes and 84 nitroimidazoles were utilised for two-dimensional QSAR modeling to retrieve their structural features essential to elicit radiosensitivity. The work comprises two parts: (i) local modeling using individual datasets; and (ii) global modeling by clubbing the three datasets. The two-dimensional descriptors were calculated using Dragon (version 7.0) software. The developed models were obtained using various feature selection techniques applied in “Small Dataset Modeling” and “Double Cross Validation” tools available from https://dtclab.webs.com/software-tools. Finally, the models were validated using stringent metrics following the Organisation for Economic Co-operation and Development (OECD) guidelines. The developed models are robust, predictive, and are useful tools to predict the radiosensitization of newly developed nitroaromatics. Furthermore, the global model was used to predict two external sets comprising 10 and 47 compounds, and the prediction ability was validated using the “Prediction Reliability Indicator” tool.

QSAR STUDY OF PYRAZOLE-UREA HYBRID COMPOUNDS AS ANTIMALARIAL AGENT VIA PROLYL-tRNA SYNTHETASE INHIBITION

Aminoacyl-tRNA synthetase is one of the emerging potential targets for various tropical diseases such as malaria. One of which, prolyl-tRNA synthetase, has known to possess diverse binding modes such as dual inhibitor with febrifugine and the recently elucidated allosteric site. A recent study showed that pyrazole-urea hybrid analogues are a potential selective inhibitor agent against this enzyme. Here, we have created a quantitative structure-activity relationship (QSAR) model using multiple linear regression (MLR) and a partial least square (PLS) regression approach. Due to the limited amount of available data, the double cross-validation (DCV) method was employed to overcome a bias which resulted due to some data points being held out for validation. Several optimizations were performed to obtain the best result, namely different data splitting strategies (random and rational) and different correlation threshold values (0.8 and 0.9). Finally, the two best equations obtained from MLR and PLS approaches were evaluated by checking their applicability domain. It is found that the MLR output model yielded the best result, which indicated the significance of Geary autocorrelation and CATS-2D-based descriptors in representing the correlation between pyrazole-urea hybrid analogues and their enzymatic activity.

Identifying Predictors of Opioid Overdose Death at a Neighborhood Level With Machine Learning.

Predictors of opioid overdose death in neighborhoods are important to identify, both to understand characteristics of high-risk areas and to prioritize limited prevention and intervention resources. Machine learning methods could serve as a valuable tool for identifying neighborhood-level predictors. We examined statewide data on opioid overdose death from Rhode Island (log-transformed rates for 2016-2019) and 203 covariates from the American Community Survey for 742 US Census block groups. The analysis included a least absolute shrinkage and selection operator (LASSO) algorithm followed by variable importance rankings from a random forest algorithm. We employed double cross-validation, with 10 folds in the inner loop to train the model and 4 outer folds to assess predictive performance. The ranked variables included a range of dimensions of socioeconomic status, including education, income and wealth, residential stability, race/ethnicity, social isolation, and occupational status. The R2 value of the model on testing data was 0.17. While many predictors of overdose death were in established domains (education, income, occupation), we also identified novel domains (residential stability, racial/ethnic distribution, and social isolation). Predictive modeling with machine learning can identify new neighborhood-level predictors of overdose in the continually evolving opioid epidemic and anticipate the neighborhoods at high risk of overdose mortality.

Origin of indigo colorants revealed by ion mobility spectrometry coupled to mass spectrometry followed by supervised classification

Identification of colorants is a key theme in art history and restoration. Indigo, the famous blue dye and pigment, was historically produced from four biological sources cultivated in specific parts of the world: woad in Europe, true indigo in Indian region, anil in South and Central America, and Japanese indigo in Eastern Asia. Therefore, identification of the biological source of indigo could reveal the provenance of pigments and dyes used in historical artworks, assist evaluation of their age, and help to trace ancient trade routes. To date, none of the previous reports achieved confident identification of indigo sources. This is the first systematic comprehensive untargeted workflow for discrimination of indigo colorants from different biological sources. Besides the new analytical procedure based on flow injection analysis and high-resolution mass spectrometry coupled to ion mobility spectrometry, the original multi-method screening scheme for feature selection, classification model training, and validation is presented. Reference indigo pigments were analyzed neat or embedded into appropriate binders to simulate oil painting or egg tempera mural. Combinations of eight feature selection methods (information gain, information gain ratio, Gini decrease, ANOVA, Chi squared, ReliefF, fast correlation based filter, and significance analysis of microarrays) and five classification algorithms (k-nearest neighbors, support-vector machines, random forest, neural network, and naïve Bayes) were evaluated via double (nested) cross-validation. The best performance was achieved with random forest across all feature selection methods (average classification accuracy 94%, negative ionization mode data). Tentative substance classes were proposed for the most important features based on their mass fragmentation patterns. A methylated flavonoid (diosmetin) and an indole alkaloid (2,2′-biindolyl) were identified in woad and true indigo pigments, respectively, and confirmed by chromatography. Plant sources of colorants were determined in historical samples of oil paintings and textiles.

Determination of Sugar, pH, and Anthocyanin Contents in Port Wine Grape Berries through Hyperspectral Imaging: An Extensive Comparison of Linear and Non-Linear Predictive Methods

This paper presents an extended comparison study between 16 different linear and non-linear regression methods to predict the sugar, pH, and anthocyanin contents of grapes through hyperspectral imaging (HIS). Despite the numerous studies on this subject that can be found in the literature, they often rely on the application of one or a very limited set of predictive methods. The literature on multivariate regression methods is quite extensive, so the analytical domain explored is too narrow to guarantee that the best solution has been found. Therefore, we developed an integrated linear and non-linear predictive analytics comparison framework (L&amp;NL-PAC), fully integrated with five preprocessing techniques and five different classes of regression methods, for an effective and robust comparison of all alternatives through a robust Monte Carlo double cross-validation stratified data splitting scheme. L&amp;NLPAC allowed for the identification of the most promising preprocessing approaches, best regression methods, and wavelengths most contributing to explaining the variability of each enological parameter for the target dataset, providing important insights for the development of precision viticulture technology, based on the HSI of grape. Overall, the results suggest that the combination of the Savitzky−Golay first derivative and ridge regression can be a good choice for the prediction of the three enological parameters.

Multiblock Analysis Applied to Fluorescence and Absorbance Spectra to Estimate Total Polyphenol Content in Extra Virgin Olive Oil.

A fast and easy methodology to estimate total polyphenol content in extra virgin olive oil was developed by applying the chemometric multiblock method sequential and orthogonalized partial least squares (SO-PLS) in order to combine front-face emission fluorescence spectra (270 nm excitation wavelength) and absorbance spectra. The hypothesis of this work stated that inner-filter effects in fluorescence spectra that would reduce the estimation performance of a single block model could be overcome by incorporating the absorbance spectral information of the compounds causing them. Different spectral preprocessing algorithms were applied. Double cross-validation with 50 iterations was implemented to improve the robustness of the obtained results. The PLSR model on the single block of fluorescence raw spectra achieved an RMSEP of 177.11 mg·kg−1 as the median value, and the complexity of the model was high, as the median value of latent variables (LVs) was eight. Multiblock SO-PLS models with pretreated fluorescence and absorbance spectra provided better performance, although artefacts could be introduced by transformation. The combination of fluorescence and absorbance raw data decreased the RMSEP median to 134.45 mg·kg−1. Moreover, the complexity of the model was greatly reduced, which contributed to an increase in robustness. The median value of LVs was three for fluorescence data and only one for absorbance data. Validation of the methodology could be addressed by further work considering a higher number of samples and a detailed composition of polyphenols.