Exploration of linear modelling techniques and their combination with multivariate adaptive regression splines to predict gastro-intestinal absorption of drugs

Impact of morphological traits and irrigation levels on fresh herbage yield of sorghum x sudangrass hybrid: Modelling data mining techniques

In this study, the effect of morphological traits on fresh herbage yield of sorghum x sudangrass hybrid plant grown in Konya province, which is the largest cereal production area in Turkey, was analyzed with some data mining methods. For this purpose, Artificial Neural Networks (ANN), Automatic Linear Model (ALM), Random Forest (RF) Algorithm and Multivariate Adaptive Regression Spline (MARS) Algorithm were used, and the prediction performances of these methods were compared. Plant height of 251.22 cm, stem diameter of 7.03 mm, fresh herbage yield of 8010.69 kg da-1, crude protein ratio of 9.09%, acid detergent fiber 33.23%, neutral detergent fiber 57.44%, acid detergent lignin 7.43%, dry matter digestibility of 63.01%, dry matter intake 2.11%, and relative feed value of 103.02 were the descriptive statistical values that were computed. Model fit statistics, including coefficient of determination (R2), adjusted R2, root of mean square error (RMSE), mean absolute percentage error (MAPE), standard deviation ratio (SD ratio), Mean Absolution Error (MAE) and Relative Absolution Error (RAE), were used to evaluate the prediction abilities of the fitted models. The MARS method was shown to be the best model for describing fresh herbage yield, with the lowest values of RMSE, MAPE, SD ratio, MAE and RAE (137.7, 1.488, 0.072, 109.718 and 0.017, respectively), as well as the highest R2 value (0.995) and adjusted R2 value (0.991). The experimental results show that the MARS algorithm is the most suitable model for predicting fresh herbage yield in sorghum x sudangrass hybrid, providing a good alternative to other data mining algorithms.

Prediction of fresh herbage yield using data mining techniques with limited plant quality parameters

The purpose of this study was to ascertain the fresh herbage yield, fertilizer dosage, and plant characteristics of the Sorghum-Sudangrass hybrid grown in arid and semi-arid regions, as well as their interrelationships. For this reason, data from the Sorghum-Sudangrass hybrid were used to assess the predictive performance of several data mining techniques, including CHAID, CART, MARS, and Bagging MARS. Plant traits were measured in Konya and Sanliurfa during 2021 and 2022. The descriptive statistical values were calculated as follows: plant height 306.27 cm, stem diameter 9.47 mm, fresh herbage yield 10852.51 kg/da, crude protein ratio 9.66%, acid detergent fiber 33.39%, neutral detergent fiber 51.85%, acid detergent lignin 9.76%, dry matter digestibility 62.88%, dry matter intake 2.34%, and relative feed value 114.68 (average values). The predictive capacities of the fitted models were assessed using model fit statistics such as the coefficient of determination (R²), adjusted R², root mean square error (RMSE), mean absolute percentage error (MAPE), standard deviation ratio (SD ratio), and Akaike Information Criterion (AIC). With the lowest values for RMSE, MAPE, SD ratio, and AIC (246, 1.926, 0.085, and 845, respectively), and the highest R² value (0.993) and adjusted R² value (0.989), the MARS algorithm was determined to be the best model for characterizing fresh herbage yield. As a solid alternative to other data mining techniques, the MARS algorithm was shown to be the most appropriate model for forecasting fresh herbage production.

Investigation of the Relationships between Coat Colour, Sex, and Morphological Characteristics in Donkeys Using Data Mining Algorithms

This study was carried out in order to determine the morphological characteristics, body coat colour distribution, and body dimensions of donkeys raised in Turkey, as well as to determine the relationships between these factors. For this reason, the predictive performance of various machine learning algorithms (i.e., CHAID, Random Forest, ALM, MARS, and Bagging MARS) were compared, utilising the biometric data of donkeys. In particular, mean measurements were taken from a total of 371 donkeys (252 male and 119 female) with descriptive statistical values as follows: height at withers, 100.7 cm; rump height, 103.1 cm; body length, 103.8 cm; chest circumference, 112.8 cm; chest depth, 45.7 cm; chest width, 29.1 cm; front shin circumference, 13.5 cm; head length, 55 cm; and ear length, 22 cm. The body colour distribution of the donkeys considered in this study was calculated as 39.35% grey, 19.95% white, 21.83% black, and 18.87% brown. Model fit statistics, including the coefficient of determination (R²), mean square error, root-mean-square error (RMSE), mean absolute percentage error (MAPE), and standard deviation ratio (SD ratio), were calculated to measure the predictive ability of the fitted models. The MARS algorithm was found to be the best model for defining the body length of donkeys, with the highest R² value (0.916) and the lowest RMSE, MAPE, and SD ratio values (2.173, 1.615, and 0.291, respectively). The experimental results indicate that the most suitable model is the MARS algorithm, which provides a good alternative to other data mining algorithms for predicting the body length of donkeys.

Genomic prediction through machine learning and neural networks for traits with epistasis

•

Performance of machine learning and neural netowrks in Genomic analysis.

•

Heritability and QTL number impacts on performance machine learning methods.

•

Machine learning models in genomic analyses.

•

Neural networks can present better performance for complex quantitative traits.

Genomic wide selection (GWS) is one contributions of molecular genetics to breeding. Machine learning (ML) and artificial neural networks (ANN) methods are non-parameterized and can develop more accurate and parsimonious models for GWS analysis. Multivariate Adaptive Regression Splines (MARS) is considered one of the most flexible ML methods, automatically modeling nonlinearities and interactions of the predictor variables. This study aimed to evaluate and compare methods based on ANN, ML, including MARS, and G-BLUP through GWS. An F2 population formed by 1000 individuals and genotyped for 4010 SNP markers and twelve traits from a model considering epistatic effect, with QTL numbers ranging from eight to 480 and heritability (h2) of 0.3, 0.5 or 0.8 were simulated. Variation in heritability and number of QTL impacts the performance of methods. About quantitative traits (40, 80, 120, 240, and 480 QTLs) was observed highest R2 to Radial Base Network (RBF) and G-BLUP, followed by Random Forest (RF), Bagging (BA), and Boosting (BO). RF and BA also showed better results for traits to h2 of 0.3 with R2 values 16.51% and 16.30%, respectively, while MARS methods showed better results for oligogenic traits with R2 values ranging from 39,12 % to 43,20 % in h2 of 0.5 and from 59.92% to 78,56% in h2 of 0.8. Non-additive MARS methods also showed high R2 for traits with high heritability and 240 QTLs or more. ANN and ML methods are powerful tools to predict genetic values in traits with epistatic effect, for different degrees of heritability and QTL numbers.

Generating accurate in silico predictions of acute aquatic toxicity for a range of organic chemicals: Towards similarity-based machine learning methods

There has been an increase in the use of non-animal approaches, such as in silico and/or in vitro methods, for assessing the risks of hazardous chemicals. A number of machine learning algorithms link molecular descriptors that interpret chemical structural properties with their biological activity. These computer-aided methods encounter several challenges, the most significant being the heterogeneity of datasets; more efficient and inclusive computational methods that are able to process large and heterogeneous chemical datasets are needed. In this context, this study verifies the utility of similarity-based machine learning methods in predicting the acute aquatic toxicity of diverse organic chemicals on Daphnia magna and Oryzias latipes. Two similarity-based methods were tested that employ a limited training dataset, most similar to a given fitting point, instead of using the entire dataset that encompasses a wide range of chemicals. The kernel-weighted local polynomial approach had a number of advantages over the distance-weighted k-nearest neighbor (k-NN) algorithm. The results highlight the importance of lipophilicity, electrophilic reactivity, molecular polarizability, and size in determining acute toxicity. The rigorous model validation ensures that this approach is an important tool for estimating toxicity in new or untested chemicals.

In Silico Prediction of Intestinal Permeability by Hierarchical Support Vector Regression

The vast majority of marketed drugs are orally administrated. As such, drug absorption is one of the important drug metabolism and pharmacokinetics parameters that should be assessed in the process of drug discovery and development. A nonlinear quantitative structure-activity relationship (QSAR) model was constructed in this investigation using the novel machine learning-based hierarchical support vector regression (HSVR) scheme to render the extremely complicated relationships between descriptors and intestinal permeability that can take place through various passive diffusion and carrier-mediated active transport routes. The predictions by HSVR were found to be in good agreement with the observed values for the molecules in the training set (n = 53, r² = 0.93, q CV 2 = 0.84, RMSE = 0.17, s = 0.08), test set (n = 13, q² = 0.75-0.89, RMSE = 0.26, s = 0.14), and even outlier set (n = 8, q² = 0.78-0.92, RMSE = 0.19, s = 0.09). The built HSVR model consistently met the most stringent criteria when subjected to various statistical assessments. A mock test also assured the predictivity of HSVR. Consequently, this HSVR model can be adopted to facilitate drug discovery and development.

Predicting the Absorption Potential of Chemical Compounds Through a Deep Learning Approach

The human colorectal carcinoma cell line (Caco-2) is a commonly used in-vitro test that predicts the absorption potential of orally administered drugs. In-silico prediction methods, based on the Caco-2 assay data, may increase the effectiveness of the high-throughput screening of new drug candidates. However, previously developed in-silico models that predict the Caco-2 cellular permeability of chemical compounds use handcrafted features that may be dataset-specific and induce over-fitting problems. Deep Neural Network (DNN) generates high-level features based on non-linear transformations for raw features, which provides high discriminant power and, therefore, creates a good generalized model. We present a DNN-based binary Caco-2 permeability classifier. Our model was constructed based on 663 chemical compounds with in-vitro Caco-2 apparent permeability data. Two hundred nine molecular descriptors are used for generating the high-level features during DNN model generation. Dropout regularization is applied to solve the over-fitting problem and the non-linear activation. The Rectified Linear Unit (ReLU) is adopted to reduce the vanishing gradient problem. The results demonstrate that the high-level features generated by the DNN are more robust than handcrafted features for predicting the cellular permeability of structurally diverse chemical compounds in Caco-2 cell lines.

Modifications of the chemical structure of phenolics differentially affect physiological activities in pulvinar cells of Mimosa pudica L. II. Influence of various molecular properties in relation to membrane transport

Early prediction of compound absorption by cells is of considerable importance in the building of an integrated scheme describing the impact of a compound on intracellular biological processes. In this scope, we study the structure-activity relationships of several benzoic acid-related phenolics which are involved in many plant biological phenomena (growth, flowering, allelopathy, defense processes). Using the partial least squares (PLS) regression method, the impact of molecular descriptors that have been shown to play an important role concerning the uptake of pharmacologically active compounds by animal cells was analyzed in terms of the modification of membrane potential, variations in proton flux, and inhibition of the osmocontractile reaction of pulvinar cells of Mimosa pudica leaves. The hydrogen bond donors (HBD) and hydrogen bond acceptors (HBA), polar surface area (PSA), halogen ratio (Hal ratio), number of rotatable bonds (FRB), molar volume (MV), molecular weight (MW), and molar refractivity (MR) were considered in addition to two physicochemical properties (logD and the amount of non-dissociated form in relation to pKa). HBD + HBA and PSA predominantly impacted the three biological processes compared to the other descriptors. The coefficient of determination in the quantitative structure-activity relationship (QSAR) models indicated that a major part of the observed seismonasty inhibition and proton flux modification can be explained by the impact of these descriptors, whereas this was not the case for membrane potential variations. These results indicate that the transmembrane transport of the compounds is a predominant component. An increasing number of implicated descriptors as the biological processes become more complex may reflect their impacts on an increasing number of sites in the cell. The determination of the most efficient effectors may lead to a practical use to improve drugs in the control of microbial attacks on plants.

The use of Bayesian nonlinear regression techniques for the modelling of the retention behaviour of volatile components of Artemisia species

The main aim of this work was to assess the ability of Bayesian multivariate adaptive regression splines (BMARS) and Bayesian radial basis function (BRBF) techniques for modelling the gas chromatographic retention indices of volatile components of Artemisia species. A diverse set of molecular descriptors was calculated and used as descriptor pool for modelling the retention indices. The ability of BMARS and BRBF techniques was explored for the selection of the most relevant descriptors and proper basis functions for modelling. The results revealed that BRBF technique is more reproducible than BMARS for modelling the retention indices and can be used as a method for variable selection and modelling in quantitative structure-property relationship (QSPR) studies. It is also concluded that the Markov chain Monte Carlo (MCMC) search engine, implemented in BRBF algorithm, is a suitable method for selecting the most important features from a vast number of them. The values of correlation between the calculated retention indices and the experimental ones for the training and prediction sets (0.935 and 0.902, respectively) revealed the prediction power of the BRBF model in estimating the retention index of volatile components of Artemisia species.

Synthesis, spectral, thermal and antimicrobial studies of some new tri metallic biologically active ceftriaxone complexes

Iron, cobalt, nickel and copper complexes of ceftriaxone were prepared in 1:3 ligand:metal ratio to examine the ligating properties of the different moieties of the drug. The complexes were found to have high percentages of coordinated water molecules. The modes of bonding were discussed depending on the infrared spectral absorption peaks of the different allowed vibrations. The Nujol mull electronic absorption spectra and the magnetic moment values indicated the Oh geometry of the metal ions in the complexes. The ESR spectra of the iron, cobalt, and copper complexes were determined and discussed. The thermal behaviors of the complexes were studied by TG and DTA techniques. The antimicrobial activities of the complexes were examined and compared to that of the ceftriaxone itself.

QSAR studies of a number of triazole antifungal alcohols

The activity of fungicide agents containing a quinazolinone ring was described using the quantitative structure-activity relationship (QSAR) model by applying it to data taken from literature. The title compounds exhibit two important types of activity against certain fungal pathogens, i.e. activity against yeast and activity against filamentous fungi. A correlation between both antifungal activities (e.g. FA(yst) and FA(ff)) and physicochemical parameters such as the logarithm of the n-octanol/water partition coefficient (log P), the polarizability (P), the global minimum energy (TE), the energy difference between the frontier molecular orbital (DELH) and the molar refractivity (MR), was established using multiple linear regression. The molecular descriptors of the antifungal agents were obtained by quantum chemical calculations combined with molecular modeling calculations. Statistical analysis shows that the antifungal activity depends mainly on the calculated partition coefficients, log P, of the compounds. Bi-parametric models reveal that antifungal activity relates linearly to log P and P.

Ionization-specific analysis of human intestinal absorption

This study presents a mechanistic QSAR analysis of human intestinal absorption of drugs and drug-like compounds using a data set of 567 %HIA values. Experimental data represent passive diffusion across intestinal membranes, and are considered to be reasonably free of carrier-mediated transport or other unwanted effects. A nonlinear model was developed relating %HIA to physicochemical properties of drugs (lipophilicity, ionization, hydrogen bonding, and molecular size). The model describes ion-specific intestinal permeability of drugs by both transcellular and paracellular routes, and also accounts for unstirred water layer effects. The obtained model was validated on two external data sets consisting of in vivo human jejunal permeability coefficients (P(eff)) and absorption rate constants (K(a)). Validation results demonstrate good predictive power of the model (RMSE = 0.35-0.45 log units for log K(a) and log P(eff)). High prediction accuracy together with clear physicochemical interpretation (log P, pK(a)) makes this model particularly suitable for use in property-based drug design.

Shuffling multivariate adaptive regression splines and adaptive neuro-fuzzy inference system as tools for QSAR study of SARS inhibitors

In this work, the inhibitory activity of pyridine N-oxide derivatives against human severe acute respiratory syndrome (SARS) is predicted in terms of quantitative structure–activity relationship (QSAR) models. These models were developed with the aid of multivariate adaptive regression spline (MARS) and adaptive neuro-fuzzy inference system (ANFIS) combined with shuffling cross-validation technique. A shuffling MARS algorithm is utilized to select the most important variables in QSAR modeling and then these variables were used as inputs of ANFIS to predict SARS inhibitory activities of pyridine N-oxide derivatives. A data set of 119 drug-like compounds was coded with over hundred calculated meaningful molecular descriptors. The best descriptors describing the inhibition mechanism were solvation connectivity index, length to breadth ratio, relative negative charge, harmonic oscillator of aromatic index, average molecular weight and total path count. These parameters are among topological, electronic, geometric, constitutional and aromaticity descriptors. The statistical parameters of R² and root mean square error (RMSE) are 0.884 and 0.359, respectively. The accuracy and robustness of shuffling MARS–ANFIS model in predicting inhibition behavior of pyridine N-oxide derivatives (pIC₅₀) was illustrated using leave-one-out and leave-multiple-out cross-validation techniques and also by Y-randomization. Comparison of the results of the proposed model with those of GA-PLS-ANFIS shows that the shuffling MARS–ANFIS model is superior and can be considered as a tool for predicting the inhibitory behavior of SARS drug-like molecules.

Quantitative structure-property relationship study on the determination of binding constant by fluorescence quenching

Models to predict binding constant (logK) to bovine serum albumin (BSA) should be very useful in the pharmaceutical industry to help speed up the design of new compounds, especially as far as pharmacokinetics is concerned. We present here an extensive list of logK binding constants for thirty-five compounds to BSA determined by florescence quenching from the literature. These data have allowed us the derivation of a quantitative structure-property relationship (QSPR) model to predict binding constants to BSA of compounds on the basis of their structure. A stepwise multiple linear regression (MLR) was performed to build the model. The statistical parameter provided by the MLR model (R = 0.9200, RMS = 0.3305) indicated satisfactory stability and predictive ability for the model. Using florescence quenching spectroscopy, we also experimentally determined the binding constants to BSA for two bioactive components in traditional Chinese medicines. Using the proposed model it was possible to predict the binding constants for each, which were in good agreement with the experimental results. This QSPR approach can contribute to a better understanding of structural factors of the compounds responsible for drug-protein interactions, and be useful in predicting the binding constants of other compounds.

Prediction of quantitative calibration factors of some organic compounds in gas chromatography

A quantitative structure-property relationship (QSPR) methodology that involves multilinear (Hansch-type) and nonlinear (radial basis function neural network (RBFNN)) approaches was performed to correlate the quantitative molar calibration factors (f(M)) of 140 organic compounds against structural factors. The statistical characteristics provided by the multiple linear model (R(2) = 0.963; RMS = 0.089; AARD = 3.86% for test set) indicated satisfactory stability and predictive ability, while the predictive ability of the RBFNN model is somewhat superior (R(2) = 0.983; RMS = 0.075; AARD = 3.19% for test set). The multilinear model provided some insight into the main structure factors that modulate the quantitative calibration factor of the investigated compounds.

Quantitative structure-property relationship study for estimation of quantitative calibration factors of some organic compounds in gas chromatography

Quantitative structure-property relationship (QSPR) models have been used to predict and explain gas chromatographic data of quantitative calibration factors (f(M)). This method allows for the prediction of quantitative calibration factors in a variety of organic compounds based on their structures alone. Stepwise multiple linear regression (MLR) and non-linear radial basis function neural network (RBFNN) were performed to build the models. The statistical characteristics provided by multiple linear model (R2=0.927, RMS=0.073; AARD=6.34% for test set) indicated satisfactory stability and predictive ability, while the predictive ability of RBFNN model is somewhat superior (R2=0.959; RMS=0.0648; AARD=4.85% for test set). This QSPR approach can contribute to a better understanding of structural factors of the compounds responsible for quantitative analysis by gas chromatography, and can be useful in predicting the quantitative calibration factors of other compounds.