A Solovev

Bio: A Solovev is an academic researcher. The author has contributed to research in topics: Fragment (logic). The author has an hindex of 1, co-authored 1 publications receiving 5 citations.

3d molecular fragment descriptors for structure-property modeling

Abstract: We report new 3D fragment descriptors to model parameters and properties of stereoisomeric molecules and conformers. New 3D fragment descriptors have been applied to discriminate between stereoisomers in predictive QSPR modeling of the standard free energy (∆G°) for the 1:1 inclusion complexation of 76 chiral guests with β-cyclodextrin (β-CD) and 40 chiral guests with 6-amino-6-deoxy-β-cyclodextrin (am-β-CD) in water at 298 K. The in-house software, mfSpace (Molecular Fragments Space), was used for QSPR modeling, generation and coding of the 3D fragment descriptors. The program implements the Singular Value Decomposition for Multiple Linear Regression analysis as machine learning method. We used ensemble modeling techniques which include the generation of many individual models, the selection of the most relevant ones and followed by their joint application to test compounds, i.e., applying a consensus model for average predictions. The models based on 2D and 3D fragment descriptors provide the best predictions in external fivefold cross-validation: root mean squared error RMSE = 1.1 kJ/mol and determination coefficient $$R_{{det}}^{2}$$Rdet2 = 0.918 (β-CD), RMSE = 0.89 kJ/mol and $$R_{{det}}^{2}$$Rdet2 = 0.910 (am-β-CD).

QSPR modeling of potentiometric sensitivity towards heavy metal ions for polymeric membrane sensors

Abstract: Potentiometric electrodes with plasticized membranes containing various ligands are widely employed as ion-selective sensors and as cross-sensitive sensors in multisensor systems. The design and testing of the appropriate ligands to make the sensors with required properties is a long and tedious process, which is not always successful. The concept of quantitative structure-property relationship (QSPR) seems to be an attractive complement to the ordinary ligand testing and design in potentiometric sensing. In this study we explore the feasibility of QSPR as a tool for in silico prediction of sensor performance of various ligands in PVC-plasticized potentiometric sensor membranes. The data on potentiometric sensitivity towards Cu2+, Zn2+, Cd2+, Pb2+ of membranes based on 35 nitrogen-containing ligands were employed for QSPR modeling. In spite of the limited dataset the derived models relating the chemical structures of the ligands with their electrochemical sensitivities have reasonable precision of sensitivity prediction with root mean squared errors RMSE around 5 mV/dec and squared determination coefficient R2det about 0.8 in external 10-fold cross-validation for zinc, cadmium and lead. This shows a good promise for further research in this area.

Predictive Models for HOMO and LUMO Energies of N-Donor Heterocycles as Ligands for Lanthanides Separation.

Abstract: Quantum chemical calculations combined with QSPR methodology reveal challenging perspectives for the solution of a number of fundamental and applied problems. In this work, we performed the PM7 and DFT calculations and QSPR modeling of HOMO and LUMO energies for polydentate N-heterocyclic ligands promising for the extraction separation of lanthanides because these values are related to the ligands selectivity in the respect to the target cations. Data for QSPR modeling comprised the PM7 calculated HOMO and LUMO energies of N-donor heterocycles, including several types of both known and virtual undescribed polydentate ligands. Ensemble modeling included various molecular fragments as descriptors and different variable selection techniques to build consensus models (CMs) on a training set of 388 ligands using external cross-validation. CMs were then verified to make predictions for two external test sets: 45 ligands (T1) that were similar to the ligands of the training set, and 1546 structures (T2), which were substantially different from the ligands of the training set. The consensus models predict well in 5-fold cross-validation (RMSEHOMO =0.097 eV, RMSELUMO =0.064 eV), and on the external test sets (T1: RMSEHOMO =0.26 eV, RMSELUMO =0.24 eV; T2: RMSEHOMO =0.26 eV, RMSELUMO =0.17 eV). An analysis of the results reveals that substituents in heteroaromatic rings of the ligands and at the amide nitrogens can deeply influence their metal binding properties.

Gaussian processes regression for cyclodextrin host-guest binding prediction

Abstract: Machine Learning (ML) techniques are becoming an integral part of rational drug design and discovery. Data-driven modeling regularly outperforms physics-based models for predicting molecular binding affinities, placing ML as a promising tool. Cyclodextrins are nano-cages used to improve the delivery of insoluble or toxic drugs. Due to chemical similarity to proteins, ML approaches could vastly profit to improve affinity prediction and enhance their carriable drug portfolio. Here we evaluate the performance of three well-known ML methods—Support Vector Regression (SVR), Gaussian Process Regression (GPR), and eXtreme Gradient Boosting (XGB)—to predict the binding affinity of cyclodextrin and known ligands. We perform hyperparameter tuning through Random Search. The results were compatible with the presented literature. We increased our previous prediction performance and present a GPR model to adjust to the data ( $$R^2$$ = 0.803) with low prediction errors (RMSE = 1.811 kJ/mol and MAE = 1.201 kJ/mol).