Theoretical Molecular Descriptors Research Articles

Prediction of diffusion coefficients in aqueous systems by machine learning models

Currently there are no accurate models for the prediction of diffusion coefficients at infinite dilution in aqueous systems. Frequently, models that work well for polar solvents often perform worse in the case of water. At the same time, experimental data of tracer diffusion coefficients are scarce and can be impractical to measure when information on this important transport property is required. In this work, machine learning models were developed to predict the tracer diffusion coefficient of any solute in water at atmospheric pressure. Several approaches were carried out to construct the model, using different types of input parameters: pure component properties and theoretical molecular descriptors, such as atom counts, structural fragments and fingerprints, computed using different sources. A database of 126 systems (1192 data points) was used for training and the best model showed a global average absolute relative deviation (AARD) of 3.92%, with a maximum deviation of 24.27% on the test set. This model uses as inputs the temperature and 195 molecular descriptors computed using the RDKit cheminformatics package, which can be automatically calculated from a molecular identifier thus making the model very simple to use. In comparison, the well-known Wilke-Chang equation provided an AARD of 13.03% in the same test set, demonstrating the improved accuracy of the proposed solution. The models developed in this work are provided at github.com/EgiChem/ml-D12-water-app.

Structure-property modeling of coumarins and coumarin-related compounds in pharmacotherapy of cancer by employing graphical topological indices.

Coumarins, a subgroup of colorless and crystalline oxygenated heterocyclic compounds originally discovered in the plant Dipteryx odorata, were the subject of a recent study investigating their quantitative structure-activity relationship (QSAR) in cancer pharmacotherapy. This study utilized graph theoretical molecular descriptors, also known as topological indices, as a numerical representation method for the chemical structures embedded in molecular graphs. These descriptors, derived from molecular graphs, play a pivotal role in quantitative structure-property relationship (QSPR) analysis. In this paper, intercorrelation between the Balban index, connective eccentric index, eccentricity connectivity index, harmonic index, hyper Zagreb index, first path Zagreb index, second path Zagreb index, Randic index, sum connectivity index, graph energy and Laplacian energy is studied on the set of molecular graphs of coumarins. It is found that the pairs of degree-based indices are highly intercorrelated. The use of these molecular descriptors in structure-boiling point modeling was analyzed. Finally, the curve-linear regression between considered molecular descriptors with physicochemical properties of coumarins and coumarin-related compounds is obtained.

Interpretable machine learning for the identification of estrogen receptor agonists, antagonists, and binders

An abnormal hormonal activity or exposure to endocrine-disrupting chemicals (EDCs) can cause endocrine system malfunction. Among the many interactions EDCs can affect is the disruption of estrogen signalling, which can lead to adverse health effects such as cancer, osteoporosis, neurodegenerative diseases, cardiovascular disease, insulin resistance, and obesity. Knowing which chemical can act as an EDC is a significant advantage and a practical necessity. New Approach Methodologies (NAM) computational models offer a quick and cost-effective solution for preliminary hazard assessment of chemicals without animal testing. Therefore, a machine learning approach was used to investigate the relationships between estrogen receptor (ER) activity and chemical structure to identify chemicals that can interact with ER. For this purpose, the consolidated in vitro assay data from ToxCast/Tox21 projects was used for developing Random Forest classification models for ER binding, agonists, and antagonists. The overall classification prediction accuracy reaches up to 82%, depending on whether the model predicted agonists, antagonists, or compounds that bind to the active site. Given the imbalance in endocrine disruption data, the derived models are good candidates for deprioritising chemicals and reducing animal testing. The interpretation of theoretical molecular descriptors of the models was consistent with the molecular interactions known in the ligand binding pocket. The estimated class probabilities enabled the analysis of the applicability domain of the developed models and the assessment of the predictions’ reliability, followed by the guidelines for interpreting prediction results. The models are openly accessible and useable at QsarDB.org (http://dx.doi.org/10.15152/QDB.259) according to the FAIR (Findable, Accessible, Interoperable, Reusable) principles.

VIBRATIONAL SPECTROSCOPIC STUDY OF PYRIDINE AND PYRIMIDINE LIGANDS COORDINATED WITH ANTIMONY (III) COMPLEXES: INSIGHTS FROM DFT CALCULATIONS

By employing the Scaled Quantum Mechanics Force Field (SQMFF) methodology, a comprehensive analysis was conducted to assign the vibrational spectra of three antimony (III) compounds, [1a-3a], that possess pyridine and pyrimidine ligands. The potential energy distribution (PED) was calculated and utilized to assign the IR spectra of the antimony (III) compounds. The theoretical frontier molecular orbital descriptors, the partial and total density of state distribution (TDOS, PDOS), molecular electronic potential surface map (MEP), nonlinear optical properties (NLO) of these complexes also were computed and investigated. The DFT/B3LYP/GEN (C, H, N, Cl: 6-31G(d,p) and Sb: LanL2DZ) level was utilized for all DFT calculations using the Gaussian 09W program. Furthermore, theoretical frontier molecular orbital descriptors, including electronegativity, chemical potential, softness, electrophilicity index, and electron affinity for six antimony (III) compounds were calculated ([1a/1b-3a/3b]). The results showed that, the ionization potential energy value of the [3a], which had the lowest experimental Leishmania activity, was also found to be the lowest among the others.

Curvilinear regression analysis of benzenoid hydrocarbons and computation of some reduced reverse degree based topological indices for hyaluronic acid-paclitaxel conjugates

Graph theoretical molecular descriptors alias topological indices are a convenient means for expressing in numerical form the chemical structure encoded in a molecular graph. The structure descriptors derived from molecular graphs are widely used in quantitative structure-property relationship (QSPR) and quantitative structure-activity relationship (QSAR) studies. The reason for introducing new indices is to obtain predictions of target properties of considered molecules that are better than the predictions obtained using already known indices. In this paper, we apply the reduced reverse degree based indices introduced in 2021 by Vignesh et al. In the QSPR analysis, we first compute the reduced reverse degree based indices for a family of benzenoid hydrocarbon molecules and then we obtain the correlation with the Physico-chemical properties of the considered molecules. We show that all the properties taken into consideration for the benzenoid hydrocarbons can be very effectively predicted by the reduced reverse degree based indices. Also, we have compared the predictive capability of reduced reverse degree based topological descriptors against 16 existing degree based indices. Further, we compute the defined reduced reverse degree based topological indices for Hyaluronic Acid-Paclitaxel Conjugates (HAP)_{n}, n ge 1.

Comparison of supercritical fluid chromatographic methods to predict the skin permeability of pharmaceutical and cosmetic compounds

Supercritical fluid chromatography (SFC) was explored as an alternative for liquid chromatography to predict the skin permeability of pharmaceutical and cosmetic compounds. Nine dissimilar stationary phases were applied to screen a test set of 58 compounds. The experimental retention factors (log k), in addition to two sets of theoretical molecular descriptors, were applied to model the skin permeability coefficient. Different modelling approaches, i.e. multiple linear regression (MLR) and partial least squares (PLS) regression, were used. In general, the MLR models performed better than the PLS models for a given descriptor set. The results obtained on a cyanopropyl (CN) column provided the best correlation with the skin permeability data. The retention factors obtained on this column were included in a simple MLR model, together with the octanol-water partition coefficient and the number of atoms (r² = 0.81, RMSEC = 0.537 or 20.5% and RMSECV = 0.580 or 22.1%). The overall best MLR model included the chromatographic descriptor from a phenyl column and 18 descriptors (r² = 0.98, RMSEC = 0.167 or 6.2% and RMSECV = 0.238 or 8.9%). This model showed a good fit, on top of very good predictive features. However, stepwise MLR models with a reduced complexity could also be determined, with the best performance parameters obtained with the CN-column based retention and eight descriptors (r² = 0.95, RMSEC = 0.282 or 10.7% and RMSECV = 0.353 or 13.4%). SFC thus provides a suitable alternative to the liquid chromatographic techniques previously applied to model the skin permeability.

In silico package models for deriving values of solute parameters in linear solvation energy relationships

ABSTRACT Environmental partitioning influences fate, exposure and ecological risks of chemicals. Linear solvation energy relationship (LSER) models may serve as efficient tools for estimating environmental partitioning parameter values that are commonly deficient for many chemicals. Nonetheless, scarcities of empirical solute parameter values of LSER models restricted the application. This study developed and evaluated in silico methods and models to derive the values, in which excess molar refraction, molar volume and logarithm of hexadecane/air partition coefficient were computed from density functional theory; dipolarity/polarizability parameter, solute H-bond acidity and basicity parameters were predicted by quantitative structure–activity relationship models developed with theoretical molecular descriptors. New LSER models on four physicochemical properties relevant with environmental partitioning (n-octanol/water partition coefficients, n-octanol/air partition coefficients, water solubilities, sub-cooled liquid vapour pressures) were constructed using the in silico solute parameter values, which exhibited comparable performance with conventional LSER models using the empirical solute parameter values. The package models for deriving the LSER solute parameter values, with advantages that they are free of instrumental determinations, may lay the foundation for high-throughput estimating environmental partition parameter values of diverse organic chemicals.

Predicting rejection of emerging contaminants through RO membrane filtration based on ANN-QSAR modeling approach: trends in molecular descriptors and structures towards rejections.

In this work, a quantitative structure-activity relationship (QSAR) study was performed on a set of emerging contaminants (ECs) to predict their rejections by reverse osmosis membrane (RO). A wide range of molecular descriptors was calculated by Dragon software for 72 ECs. The QSAR data was analyzed by an artificial neural network method (ANN), in which four out of 3000 theoretical molecular descriptors were chosen and their significance was computed based on the Garson method. The significance trends of descriptors were as follows in descending order: ESpm14u > R2e > SIC1 > EEig03d. The selected descriptors were ranked based on their importance and then an explorative study was conducted on the QSAR data to show the trends in molecular descriptors and structures toward the rejections values of ECs. The MLR algorithm was used to make a linear model and the results were compared with those of the nonlinear ANN algorithm. The comparison results revealed it is necessary to apply the ANN model to this data with non-linear properties. For the whole dataset, the correlation coefficient (R2) and residual mean squared error (RMSE) of the ANN and MLR methods were 0.9528, 6.4224; and 0.8753, 11.3400, respectively. The comparison results showed the superiority of ANN modeling in the analysis of ECs' QSAR data.

QSER modeling of half-wave oxidation potential of indolizines by theoretical descriptors

Indolizine derivatives hold essential biological functions and have been researched for hypoglycemic, antibacterial, anti-inflammatory, analgesic, and anti-tumor actions. Indolizine scaffold has intrigued conjecture and continuous attention and has become an effective parent system for generating powerful novel medication candidates. This research focused on applying the quantitative structure-electrochemistry relationship (QSER) approach to the half-wave potential (E1/2) for Indolizine derivatives using theoretical molecular descriptors. After calculating the descriptors and splitting the data into both sets, training and prediction. The QSER model was constructed using the Genetic Algorithm/Multiple Linear Regression (GA/MLR) technique, which was used to choose the optimal descriptors for the model. A four-parameter model has been established. Many assessment procedures, including cross-validation, external validation, and Y-scrambling testing, were used to assess the model's performance. Furthermore, the applicability domain (AD) was investigated using the Williams and Insubria graphs to assess the correctness of the established model's predictions. The constructed model exhibits great goodness-of-fit to experimental data, as well as high stability (R²=0.893, Q²LOO= 0.851, Q²LMO=0.843 RMSEtr= 0.052, s= 0.056). Prediction results show a good agreement with the experimental data of E1/2 (R²ext= 0.912, Q²F1= 0.883, Q²F2= 0.883, Q²F3= 0.919, CCCext= 0.942, RMSEext=0.045).

Predicting skin permeability of pharmaceutical and cosmetic compounds using retention on octadecyl, cholesterol-bonded and immobilized artificial membrane columns

In this study, the retention on three types of columns, an immobilized artificial membrane (IAM), a cholesterol-bonded and an octadecyl (C18) column, was applied for the prediction of skin permeability. The first two columns are biomimicking ones, which have certain components of the skin bound to the stationary phase, and were applied in HPLC, while the sub-2 µm C18 column was studied in UHPLC because of its fast features. Fifty-eight compounds were analyzed applying different mobile-phase compositions, with varying percentages of organic modifier on every column, to extrapolate the retention factor to a theoretically purely aqueous mobile phase (log kw). The retention factors, along with two sets of theoretical molecular descriptors, were used to model the skin permeability coefficient (log Kp) using multiple linear regression (MLR) and partial least squares (PLS) regression modelling. Although the retention factors (log k) on the IAM column showed a better correlation with the skin permeability, the overall best model was obtained by applying a stepwise MLR approach on the UHPLC parameters combined with some theoretical descriptors. This model showed a good fit, and on top has potential to accurately predict skin permeability values. Furthermore, the UHPLC method has the advantage of being fast and can thus be classified as a high-throughput approach.

Evaluating micellar liquid chromatographic methods on octadecyl particle-based and monolithic columns to predict the skin permeation of drug and cosmetic molecules

A micellar liquid chromatographic method was developed to assist in the modeling of the skin permeability of pharmaceutical and cosmetic compounds. The composition of the mobile phase was determined by means of a two-factor central composite design, after which it was tested on both a particle-based and monolithic column. The latter provided the opportunity to increase the flow rate from 1 to 8 mL/min without reaching too high backpressures. The micellar conditions allowed analyzing a large test set of compounds with diverse characteristics with just one mobile-phase composition. The obtained experimental chromatographic descriptors besides two sets of theoretical molecular descriptors were used to model the skin permeability coefficient log Kp, applying multiple linear regression and partial least squares regression approaches. The micellar method on the monolithic column provided useful models with similar or even slightly better performance parameters than the method on the particle-based column. Furthermore, a much faster analysis can be achieved when applying a flow rate of 8 mL/min, making the micellar monolithic method ideal to estimate skin permeability.

Water Activity Prediction in Sugar and Polyol Systems Using Theoretical Molecular Descriptors.

Water activity is a key factor in the development of pharmaceutical, cosmetic, and food products. In aqueous solutions of nonelectrolytes, the Norrish model provides a simple and effective way to evaluate this quantity. However, it contains a parameter, known as the Norrish constant, that must be estimated from experimental data. In this study, a new strategy is proposed for the prediction of water activity in the absence of experimental information, based on the use of theoretical molecular descriptors for characterizing the effects of a solute. This approach was applied to the evaluation of water activity in the presence of sugars (glucose, fructose, xylose, sucrose) and polyols (sorbitol, xylitol, glycerol, erythritol). The use of two descriptors related to the constitutional and connectivity properties of the solutes was first investigated. Subsequently, a new theoretical descriptor, named the global information index (G), was developed. By using this index, the water activity curves in the binary systems were reconstructed. The positive results obtained support the proposed strategy, as well as the possibility of including, in a single information index, the main molecular features of a solute that determine its effects on water activity.

QSAR Analysis of Quinazolinyl-arylurea Derivatives as Potential Anti- Cancer Agents: GA-MLR Chemometric Approach

Background: Cancer is the most common malignancy in men and women globally. The tyrosine kinases and serine/threonine kinases are essential to cell mediators for extra & intracellular signal transduction processes and play a key role in cell proliferation, differentiation, migration, metabolism, and programmed cell deaths. In this context, kinases are considered as a potential drug target for cancer therapy. Methods: In the present study, a two-dimensional (2D) quantitative structure-activity relationship (2D-QSAR) was performed to analyze anticancer activities of 28 quinazolinyl-arylurea (QZA) derivatives based on the liver (BEL-7402), stomach (MGC-803), and colon (HCC-827) cancer cell lines using multiple linear regression (MLR) analysis. It was accomplished using 2D-QSAR analysis on the available IC50 data of 28 molecules based on theoretical molecular descriptors to develop predictive models that correlate structural features of QZA derivatives to their anticancer activities. A suitable set of molecular descriptors, such as constitutional, topological, geometrical, electrostatic, and quantum-chemical descriptors were calculated to represent the structural features of compounds. The genetic algorithm (GA) method was used to identify the important molecular descriptors to build the QSAR models and used to predict the anti-cancer activities. Results and Discussion: The obtained 2D-QSAR models were vigorously validated using various statistical metrics using leave-one-out (LOO) and external test set prediction approaches. The best predictive models by MLR gave highly significant square of correlation coefficient (R2 train) values of 0.799, 0.815, and 0.779 for the training set, and the correlation coefficients (R2 test) were obtained 0.885, 0.929, and 0.774 for the test set for the liver, stomach, and colon cancer cell lines. The models also demonstrated good predictive power confirmed by the high value of cross-validated correlation coefficient Q2 value of 0.663, 0.717, and 0.671 for three different cancer cell lines. Importantly, the model's quality was judged as well based on mean absolute error (MAE) criteria, and the results were consistent with proposed limits by Golbraikh and Tropsha. Conclusion: The QSAR results of the study indicated that the proposed models were robust and free from chance correlation. This study indicated that maxHBint7, SpMax8_Bhm, and ETA_Beta_ns_d have positively contributed descriptors for anti-cancer activity in the liver, stomach, and colon cancer cell lines and a detailed mechanistic interpretation of each model revealed important structural features that were responsible for favorable or unfavorable for anticancer activity. The predictive ability of the proposed models was good and may be useful for developing more potent quinazolinyl-arylurea compounds as anti-cancer agents.

Quantitatively modeling of tetracycline photodegradation in low molecular weight organic acids under simulated sunlight irradiation

As the ubiquitous active components in aquatic environments, low molecular weight organic acids (LMWOAs) have a large influence on the environmental behaviors of contaminants. This research was focused on the effect of different LMWOAs including 11 aliphatic acids and 7 aromatic acids on the photodegradation kinetics of tetracycline (TC), and the development of quantitative structure-activity relationship (QSAR) model. Results showed that TC photodegradation in the presence of LMWOAs fitted pseudo-first-order photolysis kinetics, and the observed photolysis rate constant (kobs) varied from 0.077 to 0.331 h−1. The QSAR model was developed by partial-least-squares (PLS) with using a sequential approach with 25 theoretical molecular descriptors. Four descriptors including ELUMO-EHOMO, ELUMO, CCR and Qmax were found to mechanistically and statistically affect kobs.The high cross validated regression coefficient (Qcum2, 0.898) and high correlation coefficient (R2, 0.908) indicated significantly goodness-of-fit and high robustness of the model. The predicted and observed values with high agreement in the defined applicability domain featured accuracy and feasibility of model. This work provided a robust predictive method for estimating the TC photodegradation in the presence of different structures of LMWOAs.

Comparison of in-silico modelling and reversed-phase liquid chromatographic retention on an octadecyl silica column to predict skin permeability of pharmaceutical and cosmetic compounds

This study focuses on the in-silico modelling of the skin permeability using a test set of pharmaceutical and cosmetic compounds. Two sets of theoretical molecular descriptors, obtained from the E-Dragon and Vega ZZ software programs, were used in the models. Different linear regression methods, i.e. Multiple Linear Regression (MLR) and Partial Least Squares (PLS) regression, were applied for modelling and estimating the skin permeability. The best model was obtained using a stepwise MLR approach on the E-Dragon descriptor set. In a second step, the retention of the test set compounds was measured on a C18 column at two pH levels: pH 5.5 and pH 7. Different organic-modifier fractions were applied in the mobile phase to be able to extrapolate the retention factors to a log kw value, with kw the estimated retention factor in an aqueous mobile phase without organic modifier. Thereafter it was examined whether combining this chromatographic descriptor with the theoretical descriptors could improve the modelling of the skin permeability. The chromatographic descriptor often did not show an added value compared to the models containing only theoretical descriptors. Therefore, the in-silico models were preferred, and these models could be useful to predict the skin permeability of pharmaceutical and cosmetic compounds.

Interaction between Antifungal Isoxazolo[3,4-b]Pyridin 3(1H)-One Derivatives and Human Serum Proteins Analyzed with Biomimetic Chromatography and QSAR Approach

The development of effective, nontoxic antifungal agents is one of the most important challenges for medicinal chemistry. A series of isoxazolo [3,4-b]pyridine-3(1H)-one derivatives previously synthesized in our laboratory demonstrated promising antifungal properties. The main goal of this study was to investigate their retention behavior in a human serum proteins-high-performance liquid chromatography (HSA-HPLC) system and explore the molecular mechanism of HSA-isoxazolone interactions using a quantitative structure–retention relationship (QSRR) approach. In order to realize this goal, multiple linear regression (MLR) modeling has been performed. The proposed QSRR models presented correlation between experimentally determined lipophilicity and computational theoretical molecular descriptors derived from Dragon 7.0 (Talete, Milan, Italy) software on the affinity of isoxazolones to HSA. The calculated plasma protein binding (PreADMET software) as well as chromatographic lipophilicity (logkw) and phospholipophilicity (CHIIAM) parameters were statistically evaluated in relation to the determined experimental HAS affinities (logkHSA). The proposed model met the Tropsha et al. criteria R2 > 0.6 and Q2 > 0.5 These results indicate that the obtained model can be useful in the prediction of an affinity to HSA for isoxazolone derivatives and they can be considered as an attractive alternative to HSA-HPLC experiments.

QSAR of 1,3,5-triazine compounds towards inhibition of toxoplasmosis utilizing computed molecular descriptors

Growing attention has been focussed towards a considerable amount of experimental and theoretical study of anti toxoplasmosis compounds having inhibitory activities against toxoplasmosis . Toxoplasmosis is a dangerous disease in both urban as well as rural areas. It is caused by Toxoplasma gondii. The life cycle involves feline species such as both domestic and wild cats, and other felids such as lions, etc . There are various targets for developing anti toxoplasmosis agents. One of the most promising targets is dihydrofolate reductase (DHFR). 1,3,5-Triazine compounds have been reported to inhibit the T. gondii. But there is hardly any formulation of quantitative structure-activity relationship (QSAR) involving 1,3,5-triazine inhibitors to date. Therefore, it is our target in the present study to develop QSAR models based on the computed theoretical molecular descriptors to scale the essential features responsible for the DHFR inhibition. These screened features of the selected compounds will help to design the potent congeneric series.

Fuzzy Divisive Hierarchical Clustering of Solvents According to Their Experimentally and Theoretically Predicted Descriptors

The present study describes a simple procedure to separate into patterns of similarity a large group of solvents, 259 in total, presented by 15 specific descriptors (experimentally found and theoretically predicted physicochemical parameters). Solvent data is usually characterized by its high variability, different molecular symmetry, and spatial orientation. Methods of chemometrics can usefully be used to extract and explore accurately the information contained in such data. In this order, advanced fuzzy divisive hierarchical-clustering methods were efficiently applied in the present study of a large group of solvents using specific descriptors. The fuzzy divisive hierarchical associative-clustering algorithm provides not only a fuzzy partition of the solvents investigated, but also a fuzzy partition of descriptors considered. In this way, it is possible to identify the most specific descriptors (in terms of higher, smallest, or intermediate values) to each fuzzy partition (group) of solvents. Additionally, the partitioning performed could be interpreted with respect to the molecular symmetry. The chemometric approach used for this goal is fuzzy c-means method being a semi-supervised clustering procedure. The advantage of such a clustering process is the opportunity to achieve separation of the solvents into similarity patterns with a certain degree of membership of each solvent to a certain pattern, as well as to consider possible membership of the same object (solvent) in another cluster. Partitioning based on a hybrid approach of the theoretical molecular descriptors and experimentally obtained ones permits a more straightforward separation into groups of similarity and acceptable interpretation. It was shown that an important link between objects’ groups of similarity and similarity groups of variables is achieved. Ten classes of solvents are interpreted depending on their specific descriptors, as one of the classes includes a single object and could be interpreted as an outlier. Setting the results of this research into broader perspective, it has been shown that the fuzzy clustering approach provides a useful tool for partitioning by the variables related to the main physicochemical properties of the solvents. It gets possible to offer a simple guide for solvents recognition based on theoretically calculated or experimentally found descriptors related to the physicochemical properties of the solvents.

Principles of QSAR Modeling

At the end of her academic career, the author summarizes the main aspects of QSAR modeling, giving comments and suggestions according to her 23 years' experience in QSAR research on environmental topics. The focus is mainly on Multiple Linear Regression, particularly Ordinary Least Squares, using a Genetic Algorithm for variable selection from various theoretical molecular descriptors, but the comments can be useful also for other QSAR methods. The need for rigorous validation, also external, and for applicability domain check to guarantee predictivity and reliability of QSAR models is particularly highlighted. The commented approach is the “predictive” one, based on chemometrics, and is usefully applied to the prioritization of environmental pollutants. All the discussed points and the author's ideas are implemented in the software QSARINS, as a legacy to the QSAR community.

Improvement of quantitative structure–retention relationship models for chromatographic retention prediction of peptides applying individual local partial least squares models

In Reversed-Phase Liquid Chromatography, Quantitative Structure–Retention Relationship (QSRR) models for retention prediction of peptides can be built, starting from large sets of theoretical molecular descriptors. Good predictive QSRR models can be obtained after selecting the most informative descriptors. Reliable retention prediction may be an aid in the correct identification of proteins/peptides in proteomics and in chromatographic method development. Traditionally, global QSRR models are built, using a calibration set containing a representative range of analytes. In this study, a strategy is presented to build individual local Partial Least Squares (PLS) models for peptides, based on selected local calibration samples, most similar to the specific query peptide to be predicted. Similar local calibration peptides are selected from a possible calibration set. The calibration samples with the lowest Euclidian distances to the query peptide are considered as most similar. Two Euclidian distances are investigated as similarity parameter, (i) in the autoscaled descriptor space and, (ii) in the PLS factor space of the global calibration samples, both after variable selection by the Final Complexity Adapted Models (FCAM) method. The predictive abilities of individual local QSRR PLS models for peptides, developed with both Euclidian distances, are found significantly better than those of two global models, i.e. before and after FCAM variable selection. The predictive abilities of the local models, developed with distances calculated in the PLS factor space, were best.