A5080175337- Computational Drug Discovery Methods
- Machine Learning in Materials Science
- Analytical Chemistry and Chromatography
- Protein Structure and Dynamics
- Crystallization and Solubility Studies
- X-ray Diffraction in Crystallography
- Metabolomics and Mass Spectrometry Studies
- Free Radicals and Antioxidants
- Spectroscopy and Chemometric Analyses
- Chemical Synthesis and Analysis
- Chemical Thermodynamics and Molecular Structure
- Cholinesterase and Neurodegenerative Diseases
- Bioinformatics and Genomic Networks
- Quinazolinone synthesis and applications
- RNA and protein synthesis mechanisms
- Cell Adhesion Molecules Research
- HER2/EGFR in Cancer Research
- Phase Equilibria and Thermodynamics
- Synthesis and biological activity
- Inorganic and Organometallic Chemistry
- Inflammatory mediators and NSAID effects
- Crystallography and molecular interactions
- Neuroscience and Neuropharmacology Research
- Image Retrieval and Classification Techniques
- Click Chemistry and Applications
Institute of Molecular and Translational Medicine
2016-2024
Palacký University Olomouc
2016-2024
Sichuan University of Science and Engineering
2021
Hubei University
2021
Hubei University of Arts and Science
2021
Nankai University
2021
Xijing University
2021
National Institute of Advanced Industrial Science and Technology
2021
Philipps University of Marburg
2021
Kazan Federal University
2018
The estimation of accuracy and applicability QSAR QSPR models for biological physicochemical properties represents a critical problem. developed parameter "distance to model" (DM) is defined as metric similarity between the training test set compounds that have been subjected QSAR/QSPR modeling. In our previous work, we demonstrated utility optimal performance DM metrics based on standard deviation within an ensemble models. current study applies such analysis 30 Ames mutagenicity data were...
This work is devoted to the application of random forest approach QSAR analysis aquatic toxicity chemical compounds tested on Tetrahymena pyriformis. The simplex representation molecular structure implemented in HiT Software was used for descriptors generation a two-dimensional level. Adequate models based and RF statistical were obtained modeling set 644 compounds. Model predictivity validated two external test sets 339 110 high impact lipophilicity polarizability investigated determined....
Structure generators are widely used in de novo design studies and their performance substantially influences an outcome. Approaches based on the deep learning models conventional atom-based approaches may result invalid structures fail to address synthetic feasibility issues. On other hand, reaction-based synthetically feasible compounds but novelty diversity of generated be limited. Fragment-based can provide both better issue complexity structure was not explicitly addressed before. Here...
In this paper we offer a novel approach for the structural interpretation of QSAR models. The major advantage our developed methodology is its universality, i.e., it can be applied to any QSAR/QSPR model irrespective chemical descriptors and machine learning methods applied. This universality was achieved by using only information obtained from substructures compounds interest interpret outcomes. Reliability offered confirmed results three case studies, including end-points different types...
Abstract Interpretation of QSAR models is useful to understand the complex nature biological or physicochemical processes, guide structural optimization perform knowledge-based validation models. Highly predictive are usually and their interpretation non-trivial. This particularly true for modern neural networks. Various approaches these exist. However, it difficult evaluate compare performance applicability ever-emerging methods. Herein, we developed several benchmark data sets with...
The CACHE challenges are a series of prospective benchmarking exercises to evaluate progress in the field computational hit-finding. Here we report results inaugural challenge which 23 teams each selected up 100 commercially available compounds that they predicted would bind WDR domain Parkinson's disease target LRRK2, with no known ligand and only an apo structure PDB. lack binding data presumably low druggability is hit finding methods. Of 1955 molecules by participants Round 1 challenge,...
Abstract A new algorithm for the interpretation of Random Forest models has been developed. It allows to calculate contribution each descriptor calculated property value. In case simplex representation a molecular structure, contributions individual atoms can be calculated, and thus it becomes possible estimate influence separate fragments on investigated property. Such information used design compounds with predefined The proposed measure is not an alternative importance Breiman’s variable,...
This paper is devoted to the development of methodology for QSPR modeling mixtures and its application vapor/liquid equilibrium diagrams bubble point temperatures binary liquid mixtures. Two types special mixture descriptors based on SiRMS ISIDA approaches were developed. SiRMS-based fragment involve atoms belonging both components mixture, whereas fragments belong only one these components. The models built data set containing phase 167 represented by different combinations 67 pure liquids....
Pharmacophore modeling is a widely used strategy for finding new hit molecules. Since not all protein targets have available 3D structures, ligand-based approaches are still useful. Currently, there just few free pharmacophore tools, and these lot of restrictions, e.g., using template molecule alignment. We developed approach to representation matching which does require This can be quickly find identical pharmacophores in given set. Based on this representation, search preferably match...
This paper describes the Structural and Physico-Chemical Interpretation (SPCI) approach, which is an extension of a recently reported method for interpretation quantitative structure-activity relationship (QSAR) models. approach can efficiently be used to reveal structural motifs major physicochemical factors affecting investigated properties. Its efficacy was demonstrated both on classical Free-Wilson data set several sets with different end points (permeability blood-brain barrier,...
Abstract Docking of large compound collections becomes an important procedure to discover new chemical entities. Screening sets compounds may also occur in de novo design projects guided by molecular docking. To facilitate these processes, there is a need for automated tools capable efficiently docking number molecules using multiple computational nodes within reasonable timeframe. These should allow easy integration programs and provide user-friendly program interface support the...
The relationship between the aqueous solubility of more than two thousand eight hundred organic compounds and their structures was investigated using a QSPR approach based on Simplex Representation Molecular Structure (SiRMS). dataset consists 2537 diverse compounds. Multiple Linear Regression (MLR) Random Forest (RF) methods were used for statistical modeling at 2D level representation molecular structure. Statistical characteristics best models are quite good (MLR method: R(2) =0.85, Q(2)...
Abstract Here, we report the data visualization, analysis and modeling for a large set of 4830 S N 2 reactions rate constant which (log k ) was measured at different experimental conditions (solvent, temperature). The were encoded by one single molecular graph – Condensed Graph Reactions, allowed us to use conventional chemoinformatics techniques developed individual molecules. Thus, Matched Reaction Pairs approach suggested used analyses substituents effects on substrates nucleophiles...
Modern QSAR approaches have wide practical applications in drug discovery for designing potentially bioactive molecules. If such models are based on the use of 2D descriptors, important information contained spatial structures molecules is lost. The major problem constructing using 3D descriptors choice a putative conformation, which affects predictive performance. multi-instance (MI) learning approach considering multiple conformations model training could be reasonable solution to above...
Abstract Molecules are complex dynamic objects that can exist in different molecular forms (conformations, tautomers, stereoisomers, protonation states, etc.) and often it is not known which form responsible for observed physicochemical biological properties of a given molecule. This raises the problem selection correct machine learning modeling target properties. The same common to molecules (RNA, DNA, proteins)—long sequences where only key segments, cannot be located precisely, involved...
This article describes design, virtual screening, synthesis, and biological tests of novel αIIbβ3 antagonists, which inhibit platelet aggregation. Two types antagonists were developed: those binding either closed or open form the protein. At first step, available experimental data used to build QSAR models ligand- structure-based pharmacophore select most appropriate tool for ligand-to-protein docking. Virtual screening publicly databases (BioinfoDB, ZINC, Enamine sets) with developed...
Pharmacophore models are widely used for the identification of promising primary hits in compound large libraries. Recent studies have demonstrated that pharmacophores retrieved from protein-ligand molecular dynamic trajectories outperform a single crystal complex structure. However, number can be enormous, thus, making it computationally inefficient to use all them virtual screening. In this study, we proposed selection distinct representative by removal with identical three-dimensional...