QSAR modelling of aldehyde toxicity by means of optimisation of correlation weights of nearest neighbouring codes

In-depth understanding of molecular mechanisms of aldehyde toxicity to engineer robust Saccharomyces cerevisiae

Aldehydes are ubiquitous electrophilic compounds that ferment microorganisms including Saccharomyces cerevisiae encounter during the fermentation processes to produce food, fuels, chemicals, and pharmaceuticals. Aldehydes pose severe toxicity to the growth and metabolism of the S. cerevisiae through a variety of toxic molecular mechanisms, predominantly via damaging macromolecules and hampering the production of targeted compounds. Compounds with aldehyde functional groups are far more toxic to S. cerevisiae than all other functional classes, and toxic potency depends on physicochemical characteristics of aldehydes. The yeast synthetic biology community established a design-build-test-learn framework to develop S. cerevisiae cell factories to valorize the sustainable and renewable biomass, including the lignin-derived substrates. However, thermochemically pretreated biomass-derived substrate streams contain diverse aldehydes (e.g., glycolaldehyde and furfural), and biological conversions routes of lignocellulosic compounds consist of toxic aldehyde intermediates (e.g., formaldehyde and methylglyoxal), and some of the high-value targeted products have aldehyde functional group (e.g., vanillin and benzaldehyde). Numerous studies comprehensively characterized both single and additive effects of aldehyde toxicity via systems biology investigations, and novel molecular approaches have been discovered to overcome the aldehyde toxicity. Based on those novel approaches, researchers successfully developed synthetic yeast cell factories to convert lignocellulosic substrates to valuable products, including aldehyde compounds. In this mini-review, we highlight the salient relationship of physicochemical characteristics and molecular toxicity of aldehydes, the molecular detoxification and macromolecules protection mechanisms of aldehydes, and the advances of engineering robust S. cerevisiae against complex mixtures of aldehyde inhibitors. KEY POINTS: • We reviewed structure-activity relationships of aldehyde toxicity on S. cerevisiae. • Two-tier protection mechanisms to alleviate aldehyde toxicity are presented. • We highlighted the strategies to overcome the synergistic toxicity of aldehydes.

Ecotoxicological QSAR modeling of organic compounds against fish: Application of fragment based descriptors in feature analysis

Organic compounds (OCs) constitute an enormously large class of highly persistent and toxic chemicals widely used for various purposes throughout the world. Their increased detection in water bodies, mainly sewage treatment plants via industrial discharge, has rendered them to become a cause for ecological concern. The limited availability of experimental toxicological data has necessitated development of models that can help us identify the most hazardous and potentially toxic compounds thus prioritizing the experiments on the selected chemicals. Computational tools such as quantitative structure-activity relationship (QSAR) can be used to predict the missing data and classify the chemicals based on their acute predicted responses for existing as well as not yet synthesized chemicals. In the current study, novel, externally validated, highly robust local QSAR models for different chemical classes and moderately robust global QSAR models were developed using partial least squares (PLS) regression technique using a large dataset of 1121 OCs for the fish mortality endpoint. For feature selection, genetic algorithm along with stepwise regression was used while model validation was performed using various stringent validation criteria following the strict rules of OECD guidelines of QSAR validation. The variables included in the models were obtained from simplex representation of molecular structures (SiRMS) (Version 4.1.2.270), Dragon (Version 7.0) and PaDEL-descriptor software (Version 2.20). The final developed models were robust, externally predictive and characterized by a large chemical as well as biological domain. The predictive efficiency of the developed models was then compared with the ECOSAR tool in order to justify the applicability of the developed models in ecotoxicological predictions for organic chemicals. Better predictive efficiency of the developed QSAR models compared to the ECOSAR derived predictions signifies their applicability in early risk assessment of known as well as untested chemicals in order to design safer alternatives to the environment.

In silico prediction of pesticide aquatic toxicity with chemical category approaches

Aquatic toxicity is an important issue in pesticide development. In this study, using nine molecular fingerprints to describe pesticides, binary and ternary classification models were constructed to predict aquatic toxicity of pesticides via six machine learning methods: Naïve Bayes (NB), Artificial Neural Network (ANN), k-Nearest Neighbor (kNN), Classification Tree (CT), Random Forest (RF) and Support Vector Machine (SVM). For the binary models, local models were obtained with 829 pesticides on rainbow trout (RT) and 151 pesticides on lepomis (LP), and global models were constructed on the basis of 1258 diverse pesticides on RT and LP and 278 on other fish species. After analyzing the local binary models, we found that fish species caused influence in terms of accuracy. Considering the data size and predictive range, the 1258 pesticides were also used to build global ternary models. The best local binary models were Maccs_ANN for RT and Maccs_SVM for LP, which exhibited accuracies of 0.90 and 0.90, respectively. For global binary models, the best model was Graph_SVM with an accuracy of 0.89. Accuracy of the best global ternary model Graph_SVM was 0.81, which was a little lower than that of the best global binary model. In addition, several substructural alerts were identified including nitrobenzene, chloroalkene and nitrile, which could significantly correlate with pesticide aquatic toxicity. This study provides a useful tool for an early evaluation of pesticide aquatic toxicity in environmental risk assessment.

Modeling the toxicity of chemical pesticides in multiple test species using local and global QSTR approaches

The safety assessment processes require the toxicity data of chemicals in multiple test species and thus, emphasize the need for computational methods capable of toxicity prediction in multiple test species. Pesticides are designed toxic substances and find extensive applications worldwide. In this study, we have established local and global QSTR (quantitative structure-toxicity relationship) and ISC QSAAR (interspecies correlation quantitative structure activity-activity relationship) models for predicting the toxicities of pesticides in multiple aquatic test species using the toxicity data in crustacean (Daphnia magna, Americamysis bahia, Gammarus fasciatus, and Penaeus duorarum) and fish (Oncorhynchus mykiss and Lepomis macrochirus) species in accordance with the OECD guidelines. The ensemble learning based QSTR models (decision tree forest, DTF and decision tree boost, DTB) were constructed and validated using several statistical coefficients derived on the test data. In all the QSTR and QSAAR models, Log P was an important predictor. The constructed local, global and interspecies QSAAR models yielded high correlations (R2) of >0.941; >0.943 and >0.826, respectively between the measured and model predicted endpoint toxicity values in the test data. The performances of the local and global QSTR models were comparable. Furthermore, the chemical applicability domains of these QSTR/QSAAR models were determined using the leverage and standardization approaches. The results suggest for the appropriateness of the developed QSTR/QSAAR models to reliably predict the aquatic toxicity of structurally diverse pesticides in multiple test species and can be used for the screening and prioritization of new pesticides.

In silico prediction of chemical aquatic toxicity with chemical category approaches and substructural alerts

Aquatic toxicity is an important endpoint in the evaluation of chemically adverse effects on ecosystems.

Development of QSAR models for predicting anti-HIV-1 activity using the Monte Carlo method

Abstract The CORAL software (http://www.insilico.eu/coral/) has been examined as a tool for modeling anti-HIV-1 activity by quantitative structure — activity relationships (QSAR) for three different sets: (i) TIBO derivatives (n=82) (ii) anti-HIV-1 activity of 2-amino-6-arylsulfonylbenzonitriles and their congeners (n=64), and (iii) the measured binding affinity for fullerene-based HIV-1 PR inhibitors (n=48). A new global invariant ATOMPAIR of the molecular structure which can be calculated with the simplified molecular input line entry system (SMILES) was studied. The ATOMPAIR is an indicator of the joint presence of pairs of chemical elements (F, Cl, Br, N, O, S, and P) and three types of bonds (double covalent bond, triple covalent bond, and stereo chemical bond). Six random splits into sub-training, calibration, and test set were examined for each set. For the three aforementioned sets, the use of ATOMPAIR in the modeling process improves the predictive potential of the models for six random splits. Graphical abstract

Using support vector regression coupled with the genetic algorithm for predicting acute toxicity to the fathead minnow

The potential toxicity of chemicals may present adverse effects to the environment and human health. The quantitative structure-activity relationship (QSAR) provides a useful method for hazard assessment. In this study, we constructed a QSAR model based on a highly heterogeneous data set of 571 compounds from the US Environmental Protection Agency, for predicting acute toxicity to the fathead minnow (Pimephales promelas). An approach coupling support vector regression (SVR) with the genetic algorithm (GA) was developed to build the model. The generated QSAR model showed excellent data fitting and prediction abilities: the squared correlation coefficients (r(2)) for the training set and the test set were 0.826 and 0.802, respectively. Only eight critical descriptors, most of which are closely related to the toxicity mechanism, were chosen by GA-SVR, making the derived model readily interpretable. In summary, the successful case reported here highlights that our GA-SVR approach can be used as a general machine learning method for toxicity prediction.

CORAL: QSPR models for solubility of [C60] and [C70] fullerene derivatives

Quantitative structure-property relationships (QSPRs) between the molecular structure of [C(60)] and [C(70)] fullerene derivatives and their solubility in chlorobenzene (mg/mL) have been established by means of CORAL (CORrelations And Logic) freeware. The CORAL models are based on representation of the molecular structure by simplified molecular input line entry system (SMILES). Three random splits into the training and the external validation sets have been examined. The ranges of statistical characteristics of these models are as follows: n = 18, r (2) = 0.748-0.815, s = 15.1 -17.5 (mg/mL), F = 47-71 (training set); n = 9, r (2) = 0.806-0.936, s = 12.5-17.5 (mg/mL), F = 29-103 (validation set).

SMILES as an alternative to the graph in QSAR modelling of bee toxicity

Simplified Molecular Input Line Entry System (SMILES) nomenclature has been used as elucidating the molecular structure in construction of the quantitative structure-activity relationships (QSAR) for predicting bee toxicity. On the basis of the symbols used in the SMILES notation numerical parameters have been obtained, which are simple and fast to calculate. The method has been used to develop a QSAR model to predict toxicity of pesticides on bees. Results on a heterogeneous set of pesticides are good. Statistical characteristics of this model are: n=85, R2=0.68, s=0.82, F=180 (training set); n=20, R2=0.72, s=0.68, F=46 (test set).

Optimisation of correlation weights of SMILES invariants for modelling oral quail toxicity

The SMILES (simplified molecular input line entry system) nomenclature was used to elucidate the molecular structure in constructing the quantitative structure-property/activity relationships (QSPR/QSAR) for predicting quail toxicity after oral exposure. The presence of chemical elements in different electronic states (e.g., C, c, O, o, Cl, Br, etc.) and of different covalent bonds (i.e., -,=, and #) was used as local invariants. Combinations of different presence/absence codes for local features of the SMILES were used as global invariants. The statistical characteristics of this model are n=97, r(2)=0.755, s=0.445, F=293 (training set); n=18, r(2)=0.731, s=0.587, F=43 (test set).

Comparison of QSPR models of octanol/water partition coefficient for vitamins and non vitamins

Comparison of QSPR models for vitamins and for various substances, which are not vitamins indicates that vitamins have less number of molecular features (topologic and kinds of atomic orbitals), but, most probably, more complex mechanisms of interactions with molecules of water and octanol.

Simple Stochastic Fingerprints Towards Mathematical Modeling in Biology and Medicine 2. Unifying Markov Model for Drugs Side Effects

Most of present mathematical models for biological activity consider just the molecular structure. In the present article we pretend extending the use of Markov chain models to define novel molecular descriptors, which consider in addition other parameters like target site or biological effect. Specifically, this mathematical model takes into consideration not only the molecular structure but the specific biological system the drug affects too. Herein, a general Markov model is developed that describes 19 different drugs side effects grouped in eight affected biological systems for 178 drugs, being 270 cases finally. The data was processed by linear discriminant analysis (LDA) classifying drugs according to their specific side effects, forward stepwise was fixed as strategy for variables selection. The average percentage of good classification and number of compounds used in the training/predicting sets were 100/95.8% for endocrine manifestations, (18 out of 18)/(13 out of 14); 90.5/92.3% for gastrointestinal manifestations, (38 out of 42)/(30 out of 32); 88.5/86.5% for systemic phenomena, (23 out of 26)/(17 out of 20); 81.8/77.3% for neurological manifestations, (27 out of 33)/(19 out of 25); 81.6/86.2% for dermal manifestations, (31 out of 38)/(25 out of 29); 78.4/85.1% for cardiovascular manifestation, (29 out of 37)/(24 out of 28); 77.1/75.7% for breathing manifestations, (27 out of 35)/(20 out of 26) and 75.6/75% for psychiatric manifestations, (31 out of 41)/(23 out of 31). Additionally a back-projection analysis (BPA) was carried out for two ulcerogenic drugs to prove in structural terms the physical interpretation of the models obtained. This article develops a mathematical model that encompasses a large number of drugs side effects grouped in specifics biological systems using stochastic absolute probabilities of interaction ((A)pi(k)(j)) by the first time.

Correlation weighting of valence shells in QSAR analysis of toxicity

In the rainbow trout (Oncorhynchus mykiss), we studied the acute toxicity LC(50)-96 h of 274 organic pesticides with a wide variety of molecular structures. Optimization of correlation weights of local and global graph invariants (OCWLGI) gave quantitative structure-activity relationships (QSARs) for predicting toxicity. We used a labeled hydrogen-filled graph (LHFG) to elucidate the molecular structure. We also used the extended connectivity of zero ((0)EC(k)), first ((1)EC(k)), and second ((2)EC(k)) order, numbers of path lengths 2 (P2(k)) and 3 (P3(k)) starting from a given vertex in the LHFG, and valence shells of second order (S2(k)). S2(k) is the sum of the degree of vertices at distance 2 from a given vertex k. The presence of three-, five-, and six-member cycles and hydrogen bond indices suggested they might be used as global LHFG invariants. We applied this method to a broad set of pesticides, to predict toxicity for the trout. The best model used weighted S2(k) and global LHFG invariants. Statistical characteristics of this model are as follows: n=233, r(2)=0.7689, r(2)(pred)=0.7688, s=0.75, F=769 (training set); n=41, r(2)=0.6421, r(2)(pred)=0.4241, s=1.14, F=70 (test set).

Validation of counter propagation neural network models for predictive toxicology according to the OECD principles: a case study

The OECD has proposed five principles for validation of QSAR models used for regulatory purposes. Here we present a case study investigating how these principles can be applied to models based on Kohonen and counter propagation neural networks. The study is based on a counter propagation network model that has been built using toxicity data in fish fathead minnow for 541 compounds. The study demonstrates that most, if not all, of the OECD criteria may be met when modeling using this neural network approach.

QSAR models of quail dietary toxicity based on the graph of atomic orbitals

Graphs of atomic orbitals (GAOs) have been used to represent molecular structures. We describe rules to convert the labelled hydrogen-filled graphs (LHFGs) into GAOs. The GAO is one possible way of taking account of the structure of atoms (i.e., atomic orbitals, such as 1s(1), 2p(2) and 3d(10)) for QSPR/QSAR analyses. Optimization of correlation weights of local invariants (OCWLI) of the LHFGs and the GAOs was used to obtain a method of quail dietary toxicity modelling. Statistical characteristics of the models based on the OCWLI of GAO are better than those based on the OCWLI of the LHFGs.

QSAR models for Daphnia toxicity of pesticides based on combinations of topological parameters of molecular structures

A topological parameter is defined as an integer value of a given local or global invariant of a molecular graph. We examined three types of local graph invariants, the vertex degrees (0EC), the extended connectivity of first order (1EC), and the numbers of paths of length two (P2), as elementary invariants for construction of quantitative structure-activity relationships (QSAR). We also examined combined invariants, obtained by multiplying one of these three elementary types with another (i.e., [0EC.1EC], [0EC.P2], and [1EC.P2]), as graph invariants. Finally, global invariants were used in the QSAR analyses, codifying the presence and nature of cycles in the molecular structures under consideration. We used the correlation weights of these invariants to obtain optimal descriptors. These descriptors have been used in one-variable models to predict toxicity toward Daphnia magna for a set of pesticides. Statistical characteristics of the best model, based on the correlation weight of local topological parameters (the [0EC.P2]) together with the global topological parameters, are the following: n=220, r2=0.7822, s=0.849, F=783 (training set); n=42, r2=0.7388, s=0.941, F=113 (test set). The role of these topological parameters is discussed.

Unified drug–target interaction thermodynamic Markov model using stochastic entropies to predict multiple drugs side effects

Most of present molecular descriptors consider just the molecular structure. In the present article we pretend extending the use of Markov chain (MC) models to define novel molecular descriptors, which consider in addition other parameters like target site or toxic effect. Specifically, this molecular descriptor takes into consideration not only the molecular structure but the specific system the drug affects too. Herein, it is developed a general Markov model that describes 21 different drugs side effects grouped in 10 affected biological systems for 193 drugs, being 311 cases finally. The data were processed by linear discriminant analysis (LDA) classifying drugs according to their specific side effects, forward stepwise was fixed as strategy for variables selection. The average percentage of good classification and number of compounds used in the training/predicting sets were 92.6/91.7% for cardiovascular manifestation (25 out of 27)/(18 out of 20); 89.3/83.9% for dermal manifestations (25 out of 18)/(18 out of 21); 88.9/88.9% for endocrine manifestations (16 out of 18)/(12 out of 14); 88.9/88.2% for psychiatric manifestations (32 out of 36)/(24 out of 27); 88.5/85.6% for systemic phenomena (23 out of 26)/(17 out of 20); 85.7/91.7% for gastrointestinal manifestations (36 out of 42)/(29 out of 32); 83.3/79.2% for metabolic manifestations (15 out of 18)/(11 out of 14); 81.8/78.0% for neurological manifestations (27 out of 33)/(20 out of 25); 75.0/74.0% for hematological manifestations (36 out of 48)/(27 out of 36) and 74.3/72.8% for breathing manifestations (26 out of 35)/(19 out of 26). Finally, application of back-projection analysis (BPA) provides physic interpretation in structural terms through molecular graphics of the toxic effects predicted with these QSTR models. This article develops a mathematical model that encompasses a large number of drugs side effects grouped in specifics systems using stochastic entropies of interaction (Thetak (j)) by the first time.

Predicting multiple drugs side effects with a general drug-target interaction thermodynamic Markov model

Most of present molecular descriptors just consider the molecular structure. In the present article we pretend extending the use of Markov chain models to define novel molecular descriptors, which consider in addition to molecular structure other parameters like target site or toxic effect. Specifically, this molecular descriptor takes into consideration not only the molecular structure but the specific system the drug affects too. Herein, it is developed a general Markov model that describes 39 different drugs side effects grouped in 11 affected systems for 301 drugs, being 686 cases finally. The data was processed by linear discriminant analysis (LDA) classifying drugs according to their specific side effects, forward stepwise was fixed as strategy for variables selection. The average percentage of good classification and number of compounds used in the training/predicting sets were 100/100% for systemic phenomena (47 out of 47)/(12 out of 12) and metabolic (18 out of 18)/(5 out of 5), muscular-skeletal (23 out of 23)/(6 out of 6) and neurological manifestations (33 out of 33)/(8 out of 8); 97.6/96.7% for cardiovascular manifestation (122 out of 125)/(30 out of 31); 97.1/97.5% for breathing manifestations (34 out of 35)/(8 out of 9); 97/99.4% for gastrointestinal manifestations (159 out of 164)/(40 out of 41); 97/95% for endocrine manifestations (32 out of 33)/(7 out of 8); 96.4/94.6% for psychiatric manifestations (53 out of 55)/(13 out of 14); 95.1/99.1% for hematological manifestations (98 out of 103)/(25 out of 26) and 88/92.3% for dermal manifestations (44 out of 50)/(12 out of 13). In addition, we report preliminary experimental reversible decrease of lymphocytes differential count after administration of the antibacterial drug G-1 in mice, which coincide with a posterior probability (P%=74.91) predicted by the model. This article develops a model that encompasses a large number of side effects grouped in specific organ systems in a single stochastic framework for the first time.