MERMAID: an open source automated hit-to-lead method based on deep reinforcement learning

The hit-to-lead process makes the physicochemical properties of the hit molecules that show the desired type of activity obtained in the screening assay more drug-like. Deep learning-based molecular generative models are expected to contribute to the hit-to-lead process. The simplified molecular input line entry system (SMILES), which is a string of alphanumeric characters representing the chemical structure of a molecule, is one of the most commonly used representations of molecules, and molecular generative models based on SMILES have achieved significant success. However, in contrast to molecular graphs, during the process of generation, SMILES are not considered as valid SMILES. Further, it is quite difficult to generate molecules starting from a certain molecule, thus making it difficult to apply SMILES to the hit-to-lead process. In this study, we have developed a SMILES-based generative model that can be generated starting from a certain molecule. This method generates partial SMILES and inserts it into the original SMILES using Monte Carlo Tree Search and a Recurrent Neural Network. We validated our method using a molecule dataset obtained from the ZINC database and successfully generated molecules that were both well optimized for the objectives of the quantitative estimate of drug-likeness (QED) and penalized octanol-water partition coefficient (PLogP) optimization. The source code is available at https://github.com/sekijima-lab/mermaid.

Influence of the Physico-Chemical Properties of Model Compounds on the Mean Sizes and Retention Rate of Gliadin Nanoparticles

Vegetal proteins have emerged as appealing starting materials for the development of various drug delivery systems, and their use for obtaining polymeric nanoparticles has been profitably exploited in multidisciplinary fields. Wheat gliadin, the water-insoluble storage protein of gluten, is characterized by a great amount of hydrophobic amino acid residues and notable mucoadhesive features. This biopolymer can be easily manipulated to form colloidal carriers, films and fibers by means of bio-acceptable solvents and easy preparation procedures. In this investigation, four model compounds characterized by different octanol/water partition coefficient (logP) values were encapsulated in gliadin nanoparticles, with the aim of investigating the influence of their physico-chemical properties on the cargo features and technological characteristics of the protein nanocarriers. The results demonstrate that the chemical structure, solubility and molecular weight of the compounds used are able to dramatically modulate the mean sizes and the entrapment efficiency of gliadin nanoparticles. This demonstrates the importance of a preformulation investigation when a molecule needs to be encapsulated in this type of polymeric carrier.

Interaction of cisplatin anticancer drug with C20 bowl: DFT investigation

This study investigated the cisplatin (anticancer drug) interaction with C20 bowl and C20H10(Bowl) molecule including hydrogen-saturated with using mPW1PW91 functional. The stability of the various isomers of drug interaction with C20 bowl was investigated. The interaction energy values were estimated in these systems. Changes in the structural parameters and the frontier orbital energy and HOMO-LUMO gap values were evaluated. Charge transfer between fragments were shown with electrophilicity-based charge transfer (ECT). The Octanol–water partition coefficient (log P) and molecular volume (Vm) of these drug precursor molecules were studied. Also, Pt-C bond characterizations were illustrated using QTAIM analysis. The results showed that C20 bowl can be a promising nanocarrier for cisplatin anticancer drug.

Identification of Novel Simulants for Toxic Industrial Chemicals and Chemical Warfare Agents for Human Decontamination Studies: A Systematic Review and Categorisation of Physicochemical Characteristics

Chemical simulants have long been used in human trials of mass decontamination to determine the efficacy of decontamination interventions against more toxic agents. Until now, reliance has mostly been on individual chemicals as surrogates to specific agents (e.g., methyl salicylate for sulphur mustard). A literature review was conducted to identify chemicals that had been previously tested on human volunteers and that represent diverse physicochemical characteristics in order to create a repository for chemical simulants. Of the 171 unique chemicals identified, 78 were discounted for the risk they could pose to human volunteers, 39 were deemed suitable for use, and a further 54 were considered to be possible simulants but would require further research. Suitable simulants included both solid and liquid chemicals spanning a wide range of physicochemical properties including molecular weight, octanol/water partition coefficient, vapour pressure, and solubility. This review identifies an array of potential simulants suitable for use in human volunteer decontamination studies and is of relevance to future studies on systemic absorption and surface decontamination.

Synthesis, DFT Study and Bioactivity Evaluation of New Butanoic Acid Derivatives as Antiviral Agents

In the present work, at first, density functional theory (DFT) calculations were utilized for the molecular design of the four new butanoic acid derivatives at B3LYP/6-31+G(d) level of theory. After DFT calculations, synthesis, FT-IR, 1H NMR, and 13C NMR spectra of corresponding molecules were presented. The NBO analysis and electronic properties of the four new synthesized butanoic acid (1, 2, 3, 4) were carried out to compare their stability and reactivity. Finally, the values of octanol/water partition coefficient (miLogP), the molecular polar surface area (TPSA), the number of atoms of the molecule (natoms), the number of hydrogen bond acceptors (nON), the number of hydrogen bond donors (nOHNH), the number of violations of the Ro5 (nviolations), the number of rotatable bonds (nrotb), the molecular volume (Vm), the molecular weight (MW) and bioactivity scores were estimated and discussed.

Identification of Novel Simulants for Toxic Industrial Chemicals and Chemical Warfare Agents for Human Decontamination Studies: A Systematic Review and Categorisation of Physicochemical Characteristics

Chemical simulants have long been used in human trials of mass decontamination to determine the efficacy of decontamination interventions against more toxic agents. Until now, reliance has mostly been on individual chemicals as surrogates to specific agents (e.g. methyl salicylate for sulphur mustard). A literature review was conducted to identify chemicals that had been previously tested on human volunteers and that represent diverse physicochemical characteristics in order to create a repository for chemical simulants. Of the 171 unique chemicals identified 78 were discounted for the risk they could pose to human volunteers, 39 were deemed suitable for use and a further 54 were considered to be possible simulants but would require further research. Suitable simulants included both solid and liquid chemicals spanning a wide range of physicochemical properties including molecular weight, octanol/ water partition coefficient, vapour pressure and solubility. This review has identified an array of potential simulants suitable for use in human volunteer decontamination studies and is of relevance to future studies on systemic absorption and surface decontamination.

Determination With QSAR of Biological Activity and Relations of Between Molecular Descriptions of 5,8-Quinolinequinones Derivatives

In this study, some electronic, hydrophobic and termochemical parameters of 28 different 5,8-quinolinequinones derivatives having diversity substituents have been calculated by using DFT (B3LYP) / 6-31G (d, p) method and basis set. Relationships between different molecular descriptives have been studied with used like molecular polarizability (α), dipole moment (μ), EHOMO, ELUMO, molecular volume (Vm), ionization potential (IP), electron affinity (EA), electronegativity (χ), molecular hardness (η), molecular softness (S), electrophilic index (Ꞷ), molar refractivity (MR), octanol–water partition coefficient (logP), thermochemical properties (entropy (Se), capacity of heat (C)); as to investigate activity relationships with molecular structure. In addition, The QSAR/QSPR between molecular properties and biological activity (Anti-proliferative and Anti-inflammatory activity) has investigated, where R, R2, F and P have taken into account in order to find a statistically correct model in QSAR studies. The dependence of the electronegative parameter on both electronic and thermochemical parameters was the most correlated parameter.

Some Bounds on Bond Incident Degree Indices with Some Parameters

It is considered that there is a fascinating issue in theoretical chemistry to predict the physicochemical and structural properties of the chemical compounds in the molecular graphs. These properties of chemical compounds (boiling points, melting points, molar refraction, acentric factor, octanol-water partition coefficient, and motor octane number) are modeled by topological indices which are more applicable and well-used graph-theoretic tools for the studies of quantitative structure-property relationships (QSPRs) and quantitative structure-activity relationships (QSARs) in the subject of cheminformatics. The π -electron energy of a molecular graph was calculated by adding squares of degrees (valencies) of its vertices (nodes). This computational result, afterwards, was named the first Zagreb index, and in the field of molecular graph theory, it turned out to be a well-swotted topological index. In 2011, Vukicevic introduced the variable sum exdeg index which is famous for predicting the octanol-water partition coefficient of certain chemical compounds such as octane isomers, polyaromatic hydrocarbons (PAH), polychlorobiphenyls (PCB), and phenethylamines (Phenet). In this paper, we characterized the conjugated trees and conjugated unicyclic graphs for variable sum exdeg index in different intervals of real numbers. We also investigated the maximum value of SEIa for bicyclic graphs depending on a > 1 .