ASOptimizer: Optimizing antisense oligonucleotides through deep learning for IDO1 gene regulation

Recent studies have highlighted the effectiveness of using antisense oligonucleotides (ASOs) for cellular RNA regulation, including targets that are considered undruggable; however, manually designing optimal ASO sequences can be labor intensive and time consuming, which potentially limits their broader application. To address this challenge, we introduce a platform, the ASOptimizer, a deep-learning-based framework that efficiently designs ASOs at a low cost. This platform not only selects the most efficient mRNA target sites but also optimizes the chemical modifications for enhanced performance. Indoleamine 2,3-dioxygenase 1 (IDO1) promotes cancer survival by depleting tryptophan and producing kynurenine, leading to immunosuppression through the aryl-hydrocarbon receptor (Ahr) pathway within the tumor microenvironment. We used ASOptimizer to identify ASOs that target IDO1 mRNA as potential cancer therapeutics. Our methodology consists of two stages: sequence engineering and chemical engineering. During the sequence-engineering stage, we optimized and predicted ASO sequences that could target IDO1 mRNA efficiently. In the chemical-engineering stage, we further refined these ASOs to enhance their inhibitory activity while reducing their potential cytotoxicity. In conclusion, our research demonstrates the potential of ASOptimizer for identifying ASOs with improved efficacy and safety.

Mix-Key: graph mixup with key structures for molecular property prediction

Molecular property prediction faces the challenge of limited labeled data as it necessitates a series of specialized experiments to annotate target molecules. Data augmentation techniques can effectively address the issue of data scarcity. In recent years, Mixup has achieved significant success in traditional domains such as image processing. However, its application in molecular property prediction is relatively limited due to the irregular, non-Euclidean nature of graphs and the fact that minor variations in molecular structures can lead to alterations in their properties. To address these challenges, we propose a novel data augmentation method called Mix-Key tailored for molecular property prediction. Mix-Key aims to capture crucial features of molecular graphs, focusing separately on the molecular scaffolds and functional groups. By generating isomers that are relatively invariant to the scaffolds or functional groups, we effectively preserve the core information of molecules. Additionally, to capture interactive information between the scaffolds and functional groups while ensuring correlation between the original and augmented graphs, we introduce molecular fingerprint similarity and node similarity. Through these steps, Mix-Key determines the mixup ratio between the original graph and two isomers, thus generating more informative augmented molecular graphs. We extensively validate our approach on molecular datasets of different scales with several Graph Neural Network architectures. The results demonstrate that Mix-Key consistently outperforms other data augmentation methods in enhancing molecular property prediction on several datasets.

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.

A dual-modal graph learning framework for identifying interaction events among chemical and biotech drugs

Drug-drug interaction (DDI) identification is essential to clinical medicine and drug discovery. The two categories of drugs (i.e. chemical drugs and biotech drugs) differ remarkably in molecular properties, action mechanisms, etc. Biotech drugs are up-to-comers but highly promising in modern medicine due to higher specificity and fewer side effects. However, existing DDI prediction methods only consider chemical drugs of small molecules, not biotech drugs of large molecules. Here, we build a large-scale dual-modal graph database named CB-DB and customize a graph-based framework named CB-TIP to reason event-aware DDIs for both chemical and biotech drugs. CB-DB comprehensively integrates various interaction events and two heterogeneous kinds of molecular structures. It imports endogenous proteins founded on the fact that most drugs take effects by interacting with endogenous proteins. In the modality of molecular structure, drugs and endogenous proteins are two heterogeneous kinds of graphs, while in the modality of interaction, they are nodes connected by events (i.e. edges of different relationships). CB-TIP employs graph representation learning methods to generate drug representations from either modality and then contrastively mixes them to predict how likely an event occurs when a drug meets another in an end-to-end manner. Experiments demonstrate CB-TIP's great superiority in DDI prediction and the promising potential of uncovering novel DDIs.

Degree-Based Graph Entropy in Structure-Property Modeling

Graph entropy plays an essential role in interpreting the structural information and complexity measure of a network. Let G be a graph of order n. Suppose dG(vi) is degree of the vertex vi for each i=1,2,…,n. Now, the k-th degree-based graph entropy for G is defined as Id,k(G)=-∑i=1ndG(vi)k∑j=1ndG(vj)klogdG(vi)k∑j=1ndG(vj)k, where k is real number. The first-degree-based entropy is generated for k=1, which has been well nurtured in last few years. As ∑j=1ndG(vj)k yields the well-known graph invariant first Zagreb index, the Id,k for k=2 is worthy of investigation. We call this graph entropy as the second-degree-based entropy. The present work aims to investigate the role of Id,2 in structure property modeling of molecules.

Computation of Entropy Measures for Metal-Organic Frameworks

Entropy is a thermodynamic function used in chemistry to determine the disorder and irregularities of molecules in a specific system or process. It does this by calculating the possible configurations for each molecule. It is applicable to numerous issues in biology, inorganic and organic chemistry, and other relevant fields. Metal-organic frameworks (MOFs) are a family of molecules that have piqued the curiosity of scientists in recent years. They are extensively researched due to their prospective applications and the increasing amount of information about them. Scientists are constantly discovering novel MOFs, which results in an increasing number of representations every year. Furthermore, new applications for MOFs continue to arise, illustrating the materials' adaptability. This article investigates the characterisation of the metal-organic framework of iron(III) tetra-p-tolyl porphyrin (FeTPyP) and CoBHT (CO) lattice. By constructing these structures with degree-based indices such as the K-Banhatti, redefined Zagreb, and the atom-bond sum connectivity indices, we also employ the information function to compute entropies.

DSN-DDI: an accurate and generalized framework for drug-drug interaction prediction by dual-view representation learning

Drug-drug interaction (DDI) prediction identifies interactions of drug combinations in which the adverse side effects caused by the physicochemical incompatibility have attracted much attention. Previous studies usually model drug information from single or dual views of the whole drug molecules but ignore the detailed interactions among atoms, which leads to incomplete and noisy information and limits the accuracy of DDI prediction. In this work, we propose a novel dual-view drug representation learning network for DDI prediction ('DSN-DDI'), which employs local and global representation learning modules iteratively and learns drug substructures from the single drug ('intra-view') and the drug pair ('inter-view') simultaneously. Comprehensive evaluations demonstrate that DSN-DDI significantly improved performance on DDI prediction for the existing drugs by achieving a relatively improved accuracy of 13.01% and an over 99% accuracy under the transductive setting. More importantly, DSN-DDI achieves a relatively improved accuracy of 7.07% to unseen drugs and shows the usefulness for real-world DDI applications. Finally, DSN-DDI exhibits good transferability on synergistic drug combination prediction and thus can serve as a generalized framework in the drug discovery field.

A Paradigmatic Approach to Find the Valency-Based K-Banhatti and Redefined Zagreb Entropy for Niobium Oxide and a Metal-Organic Framework

Entropy is a thermodynamic function in chemistry that reflects the randomness and disorder of molecules in a particular system or process based on the number of alternative configurations accessible to them. Distance-based entropy is used to solve a variety of difficulties in biology, chemical graph theory, organic and inorganic chemistry, and other fields. In this article, the characterization of the crystal structure of niobium oxide and a metal-organic framework is investigated. We also use the information function to compute entropies by building these structures with degree-based indices including the K-Banhatti indices, the first redefined Zagreb index, the second redefined Zagreb index, the third redefined Zagreb index, and the atom-bond sum connectivity index.

A merged molecular representation deep learning method for blood-brain barrier permeability prediction

The ability of a compound to permeate across the blood-brain barrier (BBB) is a significant factor for central nervous system drug development. Thus, for speeding up the drug discovery process, it is crucial to perform high-throughput screenings to predict the BBB permeability of the candidate compounds. Although experimental methods are capable of determining BBB permeability, they are still cost-ineffective and time-consuming. To complement the shortcomings of existing methods, we present a deep learning-based multi-model framework model, called Deep-B3, to predict the BBB permeability of candidate compounds. In Deep-B3, the samples are encoded in three kinds of features, namely molecular descriptors and fingerprints, molecular graph and simplified molecular input line entry system (SMILES) text notation. The pre-trained models were built to extract latent features from the molecular graph and SMILES. These features depicted the compounds in terms of tabular data, image and text, respectively. The validation results yielded from the independent dataset demonstrated that the performance of Deep-B3 is superior to that of the state-of-the-art models. Hence, Deep-B3 holds the potential to become a useful tool for drug development. A freely available online web-server for Deep-B3 was established at http://cbcb.cdutcm.edu.cn/deepb3/, and the source code and dataset of Deep-B3 are available at https://github.com/GreatChenLab/Deep-B3.

SSI-DDI: substructure-substructure interactions for drug-drug interaction prediction

A major concern with co-administration of different drugs is the high risk of interference between their mechanisms of action, known as adverse drug-drug interactions (DDIs), which can cause serious injuries to the organism. Although several computational methods have been proposed for identifying potential adverse DDIs, there is still room for improvement. Existing methods are not explicitly based on the knowledge that DDIs are fundamentally caused by chemical substructure interactions instead of whole drugs' chemical structures. Furthermore, most of existing methods rely on manually engineered molecular representation, which is limited by the domain expert's knowledge.We propose substructure-substructure interaction-drug-drug interaction (SSI-DDI), a deep learning framework, which operates directly on the raw molecular graph representations of drugs for richer feature extraction; and, most importantly, breaks the DDI prediction task between two drugs down to identifying pairwise interactions between their respective substructures. SSI-DDI is evaluated on real-world data and improves DDI prediction performance compared to state-of-the-art methods. Source code is freely available at https://github.com/kanz76/SSI-DDI.

Molecular Descriptors of Some Chemicals that Prevent COVID-19

BACKGROUND

Topological index is a numerical molecular descriptor that plays an important role in structure-property/structure-activity modeling. A large number of works on multiplicative degree based indices have been developed. However, no attention is paid to investigating their chemical significance. Investigation of the chemical importance of such indices is needed. The computation of topological indices for different chemical structures and networks is a current topic of interest in mathematical chemistry.

OBJECTIVE

The objective of the present work is to examine the usefulness of the multiplicative degree based indices in quantitative structure property/activity relationship modeling. In addition, we intend to compute the indices for some anti-COVID-19 chemicals.

MATERIALS AND METHODS

The regression analysis for octane data set is performed using MATLAB and Excel to check the predictability of the indices. The sensitivity test is conducted to examine the isomer discrimination ability. To study the indices for chemical structures preventing COVID-19, different combinatorial computation methods are utilized.

RESULTS AND DISCUSSION

The regression models governing the structural dependence of different properties and activities are derived. The supremacy of the indices as useful molecular descriptors compared to some well-known and most used descriptors is established. Explicit expressions of the indices for hydroxychloroquine, remdesivir (GS-5734) and theaflavin are obtained.

CONCLUSION

As the indices are shown to have remarkable efficiency in quantitative structure property/activity relationship modeling and isomer discrimination, the outcomes can predict different properties and activities of the chemicals under consideration.

On Certain Topological Indices of Three-Layered Single-Walled Titania Nanosheets

AIM AND OBJECTIVE

Nanostructures are objects whose sizes are between microscopic and molecular. The most significant of these new elements are carbon nanotubes. These elements have extraordinary microelectronic properties and many other exclusive physiognomies. Recently, researchers have given the attention to the mathematical properties of these materials. The aim and objective of this research article is to investigate the most important molecular descriptors namely Wiener, edge-Wiener, vertex-edge-Wiener, vertex-Szeged, edge-Szeged, edge-vertex-Szeged, total-Szeged, PI, Schultz, Gutman, Mostar, edge-Mostar, and total-Mostar indices of three-layered single-walled titania nanosheets. By computing these topological indices, materials science researchers can have a better understanding of structural and physical properties of titania nanosheets, and thereby more easily synthesizing new variants of titania nanosheets with more amenable physicochemical properties.

METHODS

The cut method turned out to be extremely handy when dealing with distance-based graph invariants which are in turn among the central concepts of chemical graph theory. In this method, we use the Djokovic ́-Winkler relation to find the suitable edge cuts to leave the graph into exactly two components. Based on the graph theoretical measures of the components, we obtain the desired topological indices by mathematical computations.

RESULTS

In this paper, distance-based indices for three-layered single-walled titania nanosheets were investigated and given the exact expressions for various dimensions of three-layered single-walled titania nanosheets. These indices may be useful in synthesizing new variants of titania nanosheets and the computed topological indices play an important role in studies of Quantitative structure-activity relationship (QSAR) and Quantitative structure-property relationship (QSPR).

CONCLUSION

In this paper, we have obtained the closed expressions of several distance-based topological indices of three-layered single-walled titania nanosheet TNS_3 [m,n] molecular graph for the cases m≥ n and m < n. The graphical validations for the computed indices are done and we observe that the Wiener types, Schultz and Gutman indices perform in a similar way whereas PI and Mostar type indices perform in the same way.

TopoMS: Comprehensive topological exploration for molecular and condensed-matter systems

We introduce TopoMS, a computational tool enabling detailed topological analysis of molecular and condensed-matter systems, including the computation of atomic volumes and charges through the quantum theory of atoms in molecules, as well as the complete molecular graph. With roots in techniques from computational topology, and using a shared-memory parallel approach, TopoMS provides scalable, numerically robust, and topologically consistent analysis. TopoMS can be used as a command-line tool or with a GUI (graphical user interface), where the latter also enables an interactive exploration of the molecular graph. This paper presents algorithmic details of TopoMS and compares it with state-of-the-art tools: Bader charge analysis v1.0 (Arnaldsson et al., 01/11/17) and molecular graph extraction using Critic2 (Otero-de-la-Roza et al., Comput. Phys. Commun. 2014, 185, 1007). TopoMS not only combines the functionality of these individual codes but also demonstrates up to 4× performance gain on a standard laptop, faster convergence to fine-grid solution, robustness against lattice bias, and topological consistency. TopoMS is released publicly under BSD License. © 2018 Wiley Periodicals, Inc.

Exploring astrobiology using in silico molecular structure generation

The origin of life is typically understood as a transition from inanimate or disorganized matter to self-organized, 'animate' matter. This transition probably took place largely in the context of organic compounds, and most approaches, to date, have focused on using the organic chemical composition of modern organisms as the main guide for understanding this process. However, it has gradually come to be appreciated that biochemistry, as we know it, occupies a minute volume of the possible organic 'chemical space'. As the majority of abiotic syntheses appear to make a large set of compounds not found in biochemistry, as well as an incomplete subset of those that are, it is possible that life began with a significantly different set of components. Chemical graph-based structure generation methods allow for exhaustive in silico enumeration of different compound types and different types of 'chemical spaces' beyond those used by biochemistry, which can be explored to help understand the types of compounds biology uses, as well as to understand the nature of abiotic synthesis, and potentially design novel types of living systems.This article is part of the themed issue 'Reconceptualizing the origins of life'.

DAMQT 2.1.0: A new version of the DAMQT package enabled with the topographical analysis of electron density and electrostatic potential in molecules

DAMQT-2.1.0 is a new version of DAMQT package which includes topographical analysis of molecular electron density (MED) and molecular electrostatic potential (MESP), such as mapping of critical points (CPs), creating molecular graphs, and atomic basins. Mapping of CPs is assisted with algorithmic determination of Euler characteristic in order to provide a necessary condition for locating all possible CPs. Apart from the mapping of CPs and determination of molecular graphs, the construction of MESP-based atomic basin is a new and exclusive feature introduced in DAMQT-2.1.0. The GUI in DAMQT provides a user-friendly interface to run the code and visualize the final outputs. MPI libraries have been implemented for all the tasks to develop the parallel version of the software. Almost linear scaling of computational time is achieved with the increasing number of processors while performing various aspects of topography. A brief discussion of molecular graph and atomic basin is provided in the current article highlighting their chemical importance. Appropriate example sets have been presented for demonstrating the functions and efficiency of the code.

PI polynomial of V-phenylenic nanotubes and nanotori

The PI polynomial of a molecular graph is defined to be the sum X^{|E(G)|−N(e)} + |V(G)|(|V(G)|+1)/2 − |E(G)| over all edges of G, where N(e) is the number of edges parallel to e. In this paper, the PI polynomial of the phenylenic nanotubes and nanotori are computed. Several open questions are also included.