Simplified Molecular Input Line Entry System Research Articles

Cholinesterase Inhibitory and In Silico Toxicity Assessment of Thirty-Four Isoquinoline Alkaloids - Berberine as the Lead Compound

Cholinesterase (ChE) inhibitors used currently in clinics for the treatment of Alzheimer's disease (AD) are the most prescribed drug class with nitrogen-containing chemical formula. Galanthamine, the latest generation anti-ChE drug, contains an isoquinoline structure. The aim of the current study was to investigate the inhibitory potential of thirty-four isoquinoline alkaloids, e.g. (-)-adlumidine, β-allocryptopine, berberine, (+)-bicuculline, (-)-bicuculline, (+)-bulbocapnine, (-)-canadine, (±)-chelidimerine, corydaldine, (±)-corydalidzine, (-)-corydalmine, (+)-cularicine, dehydrocavidine, (+)-fumariline, (-)-fumarophycine, (+)-α-hydrastine, (+)-isoboldine, 13-methylcolumbamine, (-)-norjuziphine, norsanguinarine, (-)-ophiocarpine, (-)-ophiocarpine-N-oxide, oxocularine, oxosarcocapnine, palmatine, (+)-parfumine, protopine, (+)-reticuline, sanguinarine, (+)-scoulerine, (±)-sibiricine, (±)-sibiricine acetate, (-)-sinactine, and (-)-stylopine isolated from several Fumaria (fumitory) and Corydalis species towards acetyl- (AChE) and butyrylcholinesterase (BChE) by microtiter plate assays. The alkaloids with strong ChE inhibition were proceeded to molecular docking simulations as well as in silico toxicity screening for their mutagenic capacity through VEGA QSAR (AMES test) consensus model and VEGA platform as statistical approaches. The inputs were evaluated in a simplified molecular input-line entry system (SMILES). ChE inhibition assays indicated that the highest AChE inhibition was caused by berberine (IC50: 0.72 ± 0.04 µg/mL), palmatine (IC50: 6.29 ± 0.61 µg/mL), β-allocryptopine (IC50: 10.62 ± 0.45 µg/mL), (-)-sinactine (IC50: 11.94 ± 0.44 µg/mL), and dehydrocavidine (IC50: 15.01 ± 1.87 µg/mL) as compared to that of galanthamine (IC50: 0.74 ± 0.01 µg/mL), the reference drug with isoquinoline skeleton. Less number of the tested alkaloids exhibited notable BChE inhibition. Among them, berberine (IC50: 7.67 ± 0.36 µg/mL) and (-)-corydalmine (IC50: 7.78 ± 0.38 µg/mL) displayed a stronger inhibition than that of galanthamine (IC50: 12.02 ± 0.25 µg/mL). The mutagenic activity was shown for β-allocryptopine, (+)- and (-)-bicuculline, (±)-corydalidzine, (-)-corydalmine, (+)-cularicine, (-)-fumarophycine, (-)-norjuziphine, (-)-ophiocarpine-N-oxide, (+)-scoulerine, (-)-sinactine, and (-)-stylopine by means of in silico experiments. The results obtained by molecular docking simulations of berberine, palmatine, and (-)-corydalmine suggested that the estimated free ligand-binding energies of these compounds inside the binding domains of their targets are reasonable to make them capable of establishing strong polar and nonpolar bonds with the atoms of the active site amino acids. Our findings revealed that berberine, palmatin, and (-)-corydalmine stand out as the most promising isoquinoline alkaloids in terms of ChE inhibition. Among them, berberine has displayed a robust dual inhibition against both ChEs and could be evaluated further as a lead compound for AD.

In-silico green toxicology approach toward discovering safer ligands for development of safe-by-design metal–organic frameworks

A vast variety of chemical compounds have been fabricated and commercialized, they not only result in industrial exposure during manufacturing and usage, but also have environmental impacts throughout their whole life cycle. Consequently, attempts to assess the risk of chemicals in terms of toxicology have never ceased. In-silico toxicology, also known as predictive toxicology, has advanced significantly over the last decade as a result of the drawbacks of experimental investigations. In this study, ProTox-III was applied to predict the toxicity of the ligands used for metal–organic framework (MOF) design and synthesis. Initially, 35 ligands, that have been frequently utilized for MOF synthesis and fabrication, were selected. Subsequently, canonical simplified molecular-input line-entry system (SMILES) of ligands were extracted from the PUBCHEM database and inserted into the ProTox-III online server. Ultimately, webserver outputs including LD50 and the probability of toxicological endpoints (cytotoxicity, carcinogenicity, mutagenicity, immunotoxicity, and ecotoxicity) were obtained and organized. According to retrieved LD50 data, the safest ligand was 5-hydroxyisophthalic. In contrast, the most hazardous ligand was 5-chlorobenzimidazole, with an LD50 of 8 mg/kg. Among evaluated endpoints, ecotoxicity was the most active and was detected in several imidazolate ligands. This data can open new horizons in design and development of green MOFs.

HBCVTr: an end-to-end transformer with a deep neural network hybrid model for anti-HBV and HCV activity predictor from SMILES

Hepatitis B and C viruses (HBV and HCV) are significant causes of chronic liver diseases, with approximately 350 million infections globally. To accelerate the finding of effective treatment options, we introduce HBCVTr, a novel ligand-based drug design (LBDD) method for predicting the inhibitory activity of small molecules against HBV and HCV. HBCVTr employs a hybrid model consisting of double encoders of transformers and a deep neural network to learn the relationship between small molecules’ simplified molecular-input line-entry system (SMILES) and their antiviral activity against HBV or HCV. The prediction accuracy of HBCVTr has surpassed baseline machine learning models and existing methods, with R-squared values of 0.641 and 0.721 for the HBV and HCV test sets, respectively. The trained models were successfully applied to virtual screening against 10 million compounds within 240 h, leading to the discovery of the top novel inhibitor candidates, including IJN04 for HBV and IJN12 and IJN19 for HCV. Molecular docking and dynamics simulations identified IJN04, IJN12, and IJN19 target proteins as the HBV core antigen, HCV NS5B RNA-dependent RNA polymerase, and HCV NS3/4A serine protease, respectively. Overall, HBCVTr offers a new and rapid drug discovery and development screening method targeting HBV and HCV.

S-COPHY: A deep learning model for predicting the chemical class of compounds as cosmetics or pharmaceuticals based on single 3D molecular images

Non-animal-based in vitro and in silico approaches for the safety assessment of cosmetic ingredients, recently referred to as Next Generation Risk Assessment (NGRA)/New Approach Methodologies (NAMs), are evolving rapidly as approaches to provide a basis for the regulatory acceptance of new materials. However, predictive models should be applied only to chemicals within the chemical space defined by the dataset used in generating the model. Thus, only predictions for new molecules that are relatively similar to the modeling set can considered reliable with strong confidence. In this study, we developed the S-COPHY model, which employs deep learning to classify new compounds based on their structural similarity to a large collection of pharmaceutical and cosmetic compounds. S-COPHY shows high predictive accuracy both internally and externally, and in particular, there were only a few instances where pharmaceuticals were incorrectly predicted as cosmetics. The use of deep learning enabled the automatic generation of input data from SMILES (Simplified Molecular Input Line Entry System) information, resulting in more consistent model outcomes. Furthermore, GRAD-CAM (Gradient-weighted Class Activation Map) analysis provided insights into the specific structures that contribute to the model's predictions. The potentiality of S-COPHY to identify characteristic structures associated with pharmaceutical-like activity indicates its potential value in supporting safety assessments of cosmetic ingredients. Our results indicate that the S-COPHY model is a promising approach to support decision-making in large chemical spaces, thereby contributing to the safety evaluation of cosmetic ingredients. Expansion of the model to other categories, such as pesticides, could further extend its applicability.

Automated BigSMILES conversion workflow and dataset for homopolymeric macromolecules

The simplified molecular-input line-entry system (SMILES) has been utilized in a variety of artificial intelligence analyses owing to its capability of representing chemical structures using line notation. However, its ease of representation is limited, which has led to the proposal of BigSMILES as an alternative method suitable for the representation of macromolecules. Nevertheless, research on BigSMILES remains limited due to its preprocessing requirements. Thus, this study proposes a conversion workflow of BigSMILES, focusing on its automated generation from SMILES representations of homopolymers. BigSMILES representations for 4,927,181 records are provided, thereby enabling its immediate use for various research and development applications. Our study presents detailed descriptions on a validation process to ensure the accuracy, interchangeability, and robustness of the conversion. Additionally, a systematic overview of utilized codes and functions that emphasizes their relevance in the context of BigSMILES generation are produced. This advancement is anticipated to significantly aid researchers and facilitate further studies in BigSMILES representation, including potential applications in deep learning and further extension to complex structures such as copolymers.

Completing and Balancing Database Excerpted Chemical Reactions with a Hybrid Mechanistic-Machine Learning Approach.

Computer-aided synthesis planning (CASP) development of reaction routes requires an understanding of complete reaction structures. However, most reactions in the current databases are missing reaction coparticipants. Although reaction prediction and atom mapping tools can predict major reaction participants and trace atom rearrangements in reactions, they fail to identify the missing molecules to complete reactions. This is because these approaches are data-driven models trained on the current reaction databases, which comprise incomplete reactions. In this work, a workflow was developed to tackle the reaction completion challenge. This includes a heuristic-based method to identify balanced reactions from reaction databases and complete some imbalanced reactions by adding candidate molecules. A machine learning masked language model (MLM) was trained to learn from simplified molecular input line entry system (SMILES) sentences of these completed reactions. The model predicted missing molecules for the incomplete reactions, a workflow analogous to predicting missing words in sentences. The model is promising for the prediction of small- and middle-sized missing molecules in incomplete reaction records. The workflow combining both the heuristic and machine learning methods completed more than half of the entire reaction space.

Geometry-Based Molecular Generation With Deep Constrained Variational Autoencoder.

Finding target molecules with specific chemical properties plays a decisive role in drug development. We proposed GEOM-CVAE, a constrained variational autoencoder based on geometric representation for molecular generation with specific properties, which is protein-context-dependent. In terms of machine learning, it includes continuous feature embedding encoder and molecular generation decoder. Our key contribution is to propose an efficient geometric embedding method, including the spatial structure representations of drug molecule (converting the 3-D coordinates into image) and the geometric graph representations of protein target (modeling the protein surface as a mesh). The 3-D geometric information is vital to successful molecular generation, which is different from previous molecular generative methods based on 1-D or 2-D. Our model framework generates specific molecules in two phases, by first generating special image with molecular 3-D information to learn latent representations and generating molecules with constrained condition based on geometric graph convolution for specific protein and then inputting the generated structural molecules into a parser network for obtaining Simplified Molecular Input Line Entry System (SMILES) strings. Our model achieves competitive performance that implies its potential effectiveness to enable the exploration of the vast chemical space for drug discovery.

Invalid SMILES are beneficial rather than detrimental to chemical language models

Generative machine learning models have attracted intense interest for their ability to sample novel molecules with desired chemical or biological properties. Among these, language models trained on SMILES (Simplified Molecular-Input Line-Entry System) representations have been subject to the most extensive experimental validation and have been widely adopted. However, these models have what is perceived to be a major limitation: some fraction of the SMILES strings that they generate are invalid, meaning that they cannot be decoded to a chemical structure. This perceived shortcoming has motivated a remarkably broad spectrum of work designed to mitigate the generation of invalid SMILES or correct them post hoc. Here I provide causal evidence that the ability to produce invalid outputs is not harmful but is instead beneficial to chemical language models. I show that the generation of invalid outputs provides a self-corrective mechanism that filters low-likelihood samples from the language model output. Conversely, enforcing valid outputs produces structural biases in the generated molecules, impairing distribution learning and limiting generalization to unseen chemical space. Together, these results refute the prevailing assumption that invalid SMILES are a shortcoming of chemical language models and reframe them as a feature, not a bug.

Multimodal Transformer for Property Prediction in Polymers.

In this work, we designed a multimodal transformer that combines both the Simplified Molecular Input Line Entry System (SMILES) and molecular graph representations to enhance the prediction of polymer properties. Three models with different embeddings (SMILES, SMILES + monomer, and SMILES + dimer) were employed to assess the performance of incorporating multimodal features into transformer architectures. Fine-tuning results across five properties (i.e., density, glass-transition temperature (Tg), melting temperature (Tm), volume resistivity, and conductivity) demonstrated that the multimodal transformer with both the SMILES and the dimer configuration as inputs outperformed the transformer using only SMILES across all five properties. Furthermore, our model facilitates in-depth analysis by examining attention scores, providing deeper insights into the relationship between the deep learning model and the polymer attributes. We believe that our work, shedding light on the potential of multimodal transformers in predicting polymer properties, paves a new direction for understanding and refining polymer properties.

Absorption Distribution Metabolism Excretion and Toxicity Property Prediction Utilizing a Pre-Trained Natural Language Processing Model and Its Applications in Early-Stage Drug Development.

Machine learning techniques are extensively employed in drug discovery, with a significant focus on developing QSAR models that interpret the structural information of potential drugs. In this study, the pre-trained natural language processing (NLP) model, ChemBERTa, was utilized in the drug discovery process. We proposed and evaluated four core model architectures as follows: deep neural network (DNN), encoder, concatenation (concat), and pipe. The DNN model processes physicochemical properties as input, while the encoder model leverages the simplified molecular input line entry system (SMILES) along with NLP techniques. The latter two models, concat and pipe, incorporate both SMILES and physicochemical properties, operating in parallel and with sequential manners, respectively. We collected 5238 entries from DrugBank, including their physicochemical properties and absorption, distribution, metabolism, excretion, and toxicity (ADMET) features. The models' performance was assessed by the area under the receiver operating characteristic curve (AUROC), with the DNN, encoder, concat, and pipe models achieved 62.4%, 76.0%, 74.9%, and 68.2%, respectively. In a separate test with 84 experimental microsomal stability datasets, the AUROC scores for external data were 78% for DNN, 44% for the encoder, and 50% for concat, indicating that the DNN model had superior predictive capabilities for new data. This suggests that models based on structural information may require further optimization or alternative tokenization strategies. The application of natural language processing techniques to pharmaceutical challenges has demonstrated promising results, highlighting the need for more extensive data to enhance model generalization.

Electron density-based GPT for optimization and suggestion of host-guest binders.

Here we present a machine learning model trained on electron density for the production of host-guest binders. These are read out as simplified molecular-input line-entry system (SMILES) format with >98% accuracy, enabling a complete characterization of the molecules in two dimensions. Our model generates three-dimensional representations of the electron density and electrostatic potentials of host-guest systems using a variational autoencoder, and then utilizes these representations to optimize the generation of guests via gradient descent. Finally the guests are converted to SMILES using a transformer. The successful practical application of our model to established molecular host systems, cucurbit[n]uril and metal-organic cages, resulted in the discovery of 9 previously validated guests for CB[6] and 7 unreported guests (with association constant Ka ranging from 13.5 M-1 to 5,470 M-1) and the discovery of 4 unreported guests for [Pd214]4+ (with Ka ranging from 44 M-1 to 529 M-1).

Abstract B023: Modeling of new drugs clinical trials outcome with patients’ digital twins cohorts

Abstract There is a problem of clinical trial failure, as each new drug should surpass the effectiveness of existing treatment regimens, which becomes increasingly challenging over time. Another significant issue is treating patients who have developed resistance to the current therapies. Essentially, the use of drug combinations or off-label drug use, where the indication does not match the diagnosis, is akin to an experiment, as there is insufficient data on which drug or combination to use. This work proposes an approach utilizing computer modeling of patients using gene expression and clinical data. Deep learning and generative adversarial networks are employed as modeling tools. The training data for the algorithms were sourced from publicly available databases such as TCGA, Drugbank, CCLE, and GDSC. The modeling is based on the hypothesis of similarity between patients, similarity between drugs, as well as the similarity between individual organs and patient tissues with cell lines, with similarity being computed mathematically. As a result, a patient model is created, where the input consists of drugs and their combinations, and the output provides survival probability values. These model data can be generated in any required quantity with generative adversarial networks (GAN) technology to create observation and control groups. Consequently, it becomes possible to simulate clinical trials, forecasting their outcomes, and, most importantly, optimizing the trial parameters to maximize the likelihood of success. We obtained patient data and drug information from the TCGA (The Cancer Genome Atlas) database. Biologicals were excluded, and only small molecules, including targeted therapies and chemotherapeutic agents, were retained. The dataset consisted of information on 3225 patients and the effects of 161 drugs. According to our hypothesis, the patient's profile includes three groups of factors, and each of these groups contributes approximately equally to the duration of overall survival (OS) and disease-free survival (DFS) intervals. The first group comprises clinical data such as age, diagnosis, and disease stage. The second group consists of genetic data, specifically gene expression profiles, which capture the functional characteristics of genes. The third group represents the drug or combination of drugs used for patient treatment. Therefore, it is necessary to ensure that all three groups of factors have comparable dimensions of influence. For drug data, we obtained information from the DrugBank database. We represented the drug molecules using SMILES (Simplified Molecular Input Line Entry System) notation, which were then converted into 100-dimensional vectors using embedding techniques derived from natural language processing technologies. To represent combinations of drugs, we employed vector addition of the individual drug vectors. As the outcome measure for prediction, we considered numerical values representing the durations of OS and DFS intervals for each respective task. Citation Format: Dmitrii K. Chebanov, Vsevolod A. Misyurin. Modeling of new drugs clinical trials outcome with patients’ digital twins cohorts [abstract]. In: Proceedings of the AACR Special Conference on Ovarian Cancer; 2023 Oct 5-7; Boston, Massachusetts. Philadelphia (PA): AACR; Cancer Res 2024;84(5 Suppl_2):Abstract nr B023.

Python-Based Algorithm for Estimating the Parameters of Physical Property Models for Substances Not Available in Database.

An in-house Python-based algorithm was developed using simplified molecular-input line-entry specification (SMILES) strings and a dipole moment for estimating the normal boiling point, critical properties, standard enthalpy, vapor pressure, liquid molar volume, enthalpy of vaporization, heat capacity, viscosity, thermal conductivity, and surface tension of molecules. Normal boiling point, critical properties, and standard enthalpy were estimated by using the Joback group contribution method. Vapor pressure, liquid molar volume, enthalpy of vaporization, heat capacity, and surface tension were estimated by using the Riedel model, Gunn-Yamada model, Clausius-Clapeyron equation, Joback group contribution method, and Brock-Bird model, respectively. Viscosities of liquid and gas were estimated by using the Letsou-Stiel model and the Chapman-Enskog-Brokaw model, respectively. Thermal conductivities of liquid and gas were estimated by using the Sato-Riedel model and Stiel-Thodos model, respectively. Dipole moment was calculated through molecular dynamics simulation using the MMFF94 force field, performed with Avogadro software. A case study was conducted with dihydro-2-methyl-3-furanone (DHMF), 2-furaldehyde diethyl acetal (FDA), 1,1-diethoxy-3-methyl butane (DEMB), glutathione (GSH), vitamin B5 (VITB5), homocysteine (HCYS), and O-acetyl-l-homoserine (AH), which are not present in the existing property database. Cross-validation indicated that the developed Python-based algorithm provided pure component model parameters nearly identical with those obtained with the Aspen Property Constant Estimation System (PCES) method, except for the enthalpy of vaporization. The parameters for estimating the enthalpy of vaporization using the current Python-based algorithm accurately represented the behavior of the actual substances, as determined using the Clausius-Claperyon equation. This Python-based algorithm provides a detailed and clear reference for estimating pure property parameters.

Integration of the Natural Language Processing of Structural Information Simplified Molecular-Input Line-Entry System Can Improve the In Vitro Prediction of Human Skin Sensitizers.

Natural language processing (NLP) technology has recently used to predict substance properties based on their Simplified Molecular-Input Line-Entry System (SMILES). We aimed to develop a model predicting human skin sensitizers by integrating text features derived from SMILES with in vitro test outcomes. The dataset on SMILES, physicochemical properties, in vitro tests (DPRA, KeratinoSensTM, h-CLAT, and SENS-IS assays), and human potency categories for 122 substances sourced from the Cosmetics Europe database. The ChemBERTa model was employed to analyze the SMILES of substances. The last hidden layer embedding of ChemBERTa was tested with other features. Given the modest dataset size, we trained five XGBoost models using subsets of the training data, and subsequently employed bagging to create the final model. Notably, the features computed from SMILES played a pivotal role in the model for distinguishing sensitizers and non-sensitizers. The final model demonstrated a classification accuracy of 80% and an AUC-ROC of 0.82, effectively discriminating sensitizers from non-sensitizers. Furthermore, the model exhibited an accuracy of 82% and an AUC-ROC of 0.82 in classifying strong and weak sensitizers. In summary, we demonstrated that the integration of NLP of SMILES with in vitro test results can enhance the prediction of health hazard associated with chemicals.

Enhanced Structure-Based Prediction of Chiral Stationary Phases for Chromatographic Enantioseparation from 3D Molecular Conformations.

The accurate prediction of suitable chiral stationary phases (CSPs) for resolving the enantiomers of a given compound poses a significant challenge in chiral chromatography. Previous attempts at developing machine learning models for structure-based CSP prediction have primarily relied on 1D SMILES strings [the simplified molecular-input line-entry system (SMILES) is a specification in the form of a line notation for describing the structure of chemical species using short ASCII strings] or 2D graphical representations of molecular structures and have met with only limited success. In this study, we apply the recently developed 3D molecular conformation representation learning algorithm, which uses rapid conformational analysis and point clouds of atom positions in the 3D space, enabling efficient chemical structure-based machine learning. By harnessing the power of the rapid 3D molecular representation learning and a data set comprising over 300,000 chromatographic enantioseparation records sourced from the literature, our models afford notable improvements for the chemical structure-based choice of appropriate CSP for enantioseparation, paving the way for more efficient and informed decision-making in the field of chiral chromatography.

Determination of Novel SARS-CoV-2 Inhibitors by Combination of Machine Learning and Molecular Modeling Methods.

Within the scope of the project, this study aimed to find novel inhibitors by combining computational methods. In order to design inhibitors, it was aimed to produce molecules similar to the RdRp inhibitor drug Favipiravir by using the deep learning method. For this purpose, a Trained Neural Network (TNN) was used to produce 75 molecules similar to Favipiravir by using Simplified Molecular Input Line Entry System (SMILES) representations. The binding properties of molecules to Viral RNA-dependent RNA polymerase (RdRp) were studied by using molecular docking studies. To confirm the accuracy of this method, compounds were also tested against 3CL protease (3CLpro), which is another important enzyme for the progression of SARS-CoV-2. Compounds having better binding energies and RMSD values than favipiravir were searched with similarity analysis on the ChEMBL drug database in order to find similar structures with RdRp and 3CLpro inhibitory activities. A similarity search found new 200 potential RdRp and 3CLpro inhibitors structurally similar to produced molecules, and these compounds were again evaluated for their receptor interactions with molecular docking studies. Compounds showed better interaction with RdRp protease than 3CLpro. This result presented that artificial intelligence correctly produced structures similar to favipiravir that act more specifically as RdRp inhibitors. In addition, Lipinski's rules were applied to the molecules that showed the best interaction with RdRp, and 7 compounds were determined to be potential drug candidates. Among these compounds, a Molecular Dynamic simulation study was applied for ChEMBL ID:1193133 to better understand the existence and duration of the compound in the receptor site. The results confirmed that the ChEMBL ID:1193133 compound showed good Root Mean Square Deviation (RMSD), Root Mean Square Fluctuation (RMSF), hydrogen bonding, and remaining time in the active site; therefore, it was considered that it could be active against the virus. This compound was also tested for antiviral activity, and it was determined that it did not delay viral infection, although it was cytotoxic between 5mg/mL-1.25mg/mL concentrations. However, if other compounds could be tested, it might provide a chance to obtain activity, and compounds should also be tested against the enzymes as well as the other types of viruses.

Creation of Polymer Datasets with Targeted Backbones for Screening of High-Performance Membranes for Gas Separation.

A simple approach was developed to computationally construct a polymer dataset by combining simplified molecular-input line-entry system (SMILES) strings of a targeted polymer backbone and a variety of molecular fragments. This method was used to create 14 polymer datasets by combining seven polymer backbones and molecules from two large molecular datasets (MOSES and QM9). Polymer backbones that were studied include four polydimethylsiloxane (PDMS) based backbones, poly(ethylene oxide) (PEO), poly(allyl glycidyl ether) (PAGE), and polyphosphazene (PPZ). The generated polymer datasets can be used for various cheminformatics tasks, including high-throughput screening for gas permeability and selectivity. This study utilized machine learning (ML) models to screen the polymers for CO2/CH4 and CO2/N2 gas separation using membranes. Several polymers of interest were identified. The results highlight that employing an ML model fitted to polymer selectivities leads to higher accuracy in predicting polymer selectivity compared to using the ratio of predicted permeabilities.

ChemMORT: an automatic ADMET optimization platform using deep learning and multi-objective particle swarm optimization.

Drug discovery and development constitute a laborious and costly undertaking. The success of a drug hinges not only good efficacy but also acceptable absorption, distribution, metabolism, elimination, and toxicity (ADMET) properties. Overall, up to 50% of drug development failures have been contributed from undesirable ADMET profiles. As a multiple parameter objective, the optimization of the ADMET properties is extremely challenging owing to the vast chemical space and limited human expert knowledge. In this study, a freely available platform called Chemical Molecular Optimization, Representation and Translation (ChemMORT) is developed for the optimization of multiple ADMET endpoints without the loss of potency (https://cadd.nscc-tj.cn/deploy/chemmort/). ChemMORT contains three modules: Simplified Molecular Input Line Entry System (SMILES) Encoder, Descriptor Decoder and Molecular Optimizer. The SMILES Encoder can generate the molecular representation with a 512-dimensional vector, and the Descriptor Decoder is able to translate the above representation to the corresponding molecular structure with high accuracy. Based on reversible molecular representation and particle swarm optimization strategy, the Molecular Optimizer can be used to effectively optimize undesirable ADMET properties without the loss of bioactivity, which essentially accomplishes the design of inverse QSAR. The constrained multi-objective optimization of the poly (ADP-ribose) polymerase-1 inhibitor is provided as the case to explore the utility of ChemMORT.

SG-ATT: A Sequence Graph Cross-Attention Representation Architecture for Molecular Property Prediction.

Existing formats based on the simplified molecular input line entry system (SMILES) encoding and molecular graph structure are designed to encode the complete semantic and structural information of molecules. However, the physicochemical properties of molecules are complex, and a single encoding of molecular features from SMILES sequences or molecular graph structures cannot adequately represent molecular information. Aiming to address this problem, this study proposes a sequence graph cross-attention (SG-ATT) representation architecture for a molecular property prediction model to efficiently use domain knowledge to enhance molecular graph feature encoding and combine the features of molecular SMILES sequences. The SG-ATT fuses the two-dimensional molecular features so that the current model input molecular information contains molecular structure information and semantic information. The SG-ATT was tested on nine molecular property prediction tasks. Among them, the biggest SG-ATT model performance improvement was 4.5% on the BACE dataset, and the average model performance improvement was 1.83% on the full dataset. Additionally, specific model interpretability studies were conducted to showcase the performance of the SG-ATT model on different datasets. In-depth analysis was provided through case studies of in vitro validation. Finally, network tools for molecular property prediction were developed for the use of researchers.

SolPredictor: Predicting Solubility with Residual Gated Graph Neural Network.

Computational methods play a pivotal role in the pursuit of efficient drug discovery, enabling the rapid assessment of compound properties before costly and time-consuming laboratory experiments. With the advent of technology and large data availability, machine and deep learning methods have proven efficient in predicting molecular solubility. High-precision in silico solubility prediction has revolutionized drug development by enhancing formulation design, guiding lead optimization, and predicting pharmacokinetic parameters. These benefits result in considerable cost and time savings, resulting in a more efficient and shortened drug development process. The proposed SolPredictor is designed with the aim of developing a computational model for solubility prediction. The model is based on residual graph neural network convolution (RGNN). The RGNNs were designed to capture long-range dependencies in graph-structured data. Residual connections enable information to be utilized over various layers, allowing the model to capture and preserve essential features and patterns scattered throughout the network. The two largest datasets available to date are compiled, and the model uses a simplified molecular-input line-entry system (SMILES) representation. SolPredictor uses the ten-fold split cross-validation Pearson correlation coefficient R2 0.79±0.02 and root mean square error (RMSE) 1.03±0.04. The proposed model was evaluated using five independent datasets. Error analysis, hyperparameter optimization analysis, and model explainability were used to determine the molecular features that were most valuable for prediction.