Structural Information Research Articles

Improving Crystal Property Prediction from a Multiplex Graph Perspective.

Graph neural networks (GNNs) have proven to be effective tools for the rapid and accurate prediction of crystal properties. While most existing methods focus on enriching representations of crystal structures, they do not deeply explore the characteristics of crystal graphs and leverage their intrinsic information from a data science perspective. In this work, we propose the potential multiplex crystal graph neural network (PMCGNN) for crystal property prediction. Based on the characteristics of crystal graphs, we reconstruct the crystal graph into a multiplex graph that includes two views: a global crystal graph embodying infinite potentials and a local crystal graph capturing local atomic interactions. We employ graph transformers (GTs) and message passing neural networks (MPNNs) architectures to learn the atomic representations of these two perspectives. Specifically, we augment the GT by incorporating positional encodings and structural encodings from the local crystal graph. This approach promotes interaction between the two perspectives, enabling the model to learn both node positional and graph structural information from different viewpoints through an attention mechanism. As a result, it enhances the model's ability to learn crystal representations. We conduct comprehensive experiments on the JARVIS and the Materials Project data sets for evaluation. Results show that PMCGNN presents superior performance in 9 crystal prediction tasks while maintaining reasonable computational expense.

A Machine Learning Model for Predicting the Propagation Rate Coefficient in Free-Radical Polymerization.

The propagation rate coefficient (kp) is one of the most crucial kinetic parameters in free-radical polymerization (FRP) as it directly governs the rate of polymerization and the resulting molecular weight distribution. The kp in FRP can typically be obtained through experimental measurements or quantum chemical calculations, both of which can be time consuming and resource intensive. Herein, we developed a machine learning model based solely on the structural features of monomers involved in FRP, utilizing molecular embedding and a Lasso regression algorithm to predict kp more efficiently and accurately. The result shows that the model achieves a mean absolute percentage error (MAPE) of only 5.49% in the predictions for four new monomers, which indicates that the model exhibits strong generalization capabilities and provides reliable and robust predictions. In addition, this model can accurately predict the influence of the ester side chain length of (meth)acrylates on kp, aligning well with established scientific knowledge. This approach offers a straightforward and practical model for other researchers to rapidly obtain accurate kp values by employing monomer structural information. The model is sufficiently general to apply to a wide range of (meth)acrylate and butadiene FRP monomers, thereby supporting kinetic modeling of polymerization reactions.

ChemAP: predicting drug approval with chemical structures before clinical trial phase by leveraging multi-modal embedding space and knowledge distillation

Recent studies showed that the likelihood of drug approval can be predicted with clinical data and structure information of drug using computational approaches. Predicting the likelihood of drug approval can be innovative and of high impact. However, models that leverage clinical data are applicable only in clinical stages, which is not very practical. Prioritizing drug candidates and early-stage decision-making in the de novo drug development process is crucial in pharmaceutical research to optimize resource allocation. For early-stage decision-making, we need a computational model that uses only chemical structures. This seemingly impossible task may utilize the predictive power with multi-modal features including clinical data. In this work, we introduce ChemAP (Chemical structure-based drug Approval Predictor), a novel deep learning scheme for drug approval prediction in the early-stage drug discovery phase. ChemAP aims to enhance the possibility of early-stage decision-making by enriching semantic knowledge to fill in the gap between multi-modal and single-modal chemical spaces through knowledge distillation techniques. This approach facilitates the effective construction of chemical space solely from chemical structure data, guided by multi-modal knowledge related to efficacy, such as clinical trials and patents of drugs. In this study, ChemAP achieved state-of-the-art performance, outperforming both traditional machine learning and deep learning models in drug approval prediction, with AUROC and AUPRC scores of 0.782 and 0.842 respectively on the drug approval benchmark dataset. Additionally, we demonstrated its generalizability by outperforming baseline models on a recent external dataset, which included drugs from the 2023 FDA-approved list and the 2024 clinical trial failure drug list, achieving AUROC and AUPRC scores of 0.694 and 0.851. These results demonstrate that ChemAP is an effective method in predicting drug approval only with chemical structure information of drug so that decision-making can be done at the early stages of drug development process. To the best of our knowledge, our work is the first of its kind to show that prediction of drug approval is possible only with structure information of drug by defining the chemical space of approved and unapproved drugs using deep learning technology.

Noun classes, variation, and creativity in youth language practices in Zimbabwe and Tanzania

Abstract While classical accounts of Bantu linguistics often depict noun classes as an inherent feature of nouns, more recent studies show there is greater variation in the use of noun classes, especially in terms of their discursive and pragmatic context. Youth language practices, often characterized by creativity and conscious changes, are of particular interest in this regard. This paper explores whether Bantu youth language practices exhibit conscious variation in their noun class systems and agreement. I discuss data from Bantu-based youth language practices in Southern and Eastern Africa, especially Lugha ya mtaani (Tanzania) and S’ncamtho (Zimbabwe), looking at noun classes and agreement. The paper features a short discussion of variation in noun class usage in Bantu languages in general, including insights into relevant aspects such as information structure, the use of loan words, noun classes in multilingual or translanguaging situations, and semantics. The latter is particularly important, as speakers can make conscious use of noun classes as part of contextualized meaning making processes that reflect semantic nuances and stylistic choices. The objective of this contribution is to highlight the role of semantics and conscious meaning making in the choice of grammatical devices and the manifestations of microvariation in youth language practices.

LOGICAL STYLE OF ANALYSIS OF TEXT AND OTHER INFORMATION FOR SOLVING TESTS ON GENERAL COMPETENCES

The article substantiates the effectiveness of using the logical style of analyzing text and other information for solving test tasks in conditions of time shortage. For this, a general classification of tasks was carried out, visual and other methods of their solution were selected, a generalized solution method based on the coding of available information, analysis and comparison of logical structures to identify contradictions, violations of completeness and integrity, etc.

Research of methods of modeling of mass service enterprise

Objective. The purpose of the study is to use artificial intelligence tools to optimize and model the automated organization of work at a mass service enterprise. Method. The research is based on the use of a Petri net as a modeling tool, as well as mathematical hardware that makes it possible to implement simulation models of subsystems and conduct additional research on GPSS artificial intelligence models. Result. To process the results of the modeling process, the Petri net modeling method was used and a network activity model graph was developed. The modeling process used with the Petri Net application allows such functions to be described more clearly. The principles of modeling systems based on the GPSS language and Petri nets are considered. Conclusion. It is advisable to use the language devices GPSS and SP, which provide information support for the activities of a mass service enterprise, modeling the structural organization and information flows.

Quantitative phase imaging with two in-line holograms.

Quantitative phase imaging (QPI) has emerged as a practical technique for acquiring structural information from phase objects. Digital holography can realize phase detection, but it is limited by a spatial bandwidth product or affected by the overlap of conjugate images. The phase retrieval algorithm serves as an effective tool for QPI dealing with intensity patterns. Traditional phase retrieval algorithms heavily rely on strong support constraints or high data redundancy to accurately reconstruct the sample image. However, in single-frame phase retrieval algorithms, the precise acquisition of support constraints is notably challenging. The multiple-measurement spends much time on data acquisition and is unsuitable for dynamic sample observation. In this paper, we propose a novel, to the best of our knowledge, quantitative phase imaging method that utilizes only two in-line holograms. We have developed a phase retrieval algorithm based on ptychography, which eliminates twin-image and separates illumination background. The proposed method achieves high data utilization efficiency and can be employed for dynamic imaging.

Acne-related UVA-induced facial fluorescence: An exploratory study from physiological properties to tissue structure information

UVA-induced facial fluorescence (UVAF) is recognized as an objective measurement technique to quantify the severity of acne. However, notable inconsistencies in quantitative outcomes have been observed in various studies, possibly due to the fact that different colors of fluorescence represent different pathophysiological implications. This study investigated the pathophysiological importance of UVAF color differences and improved its reliability in assessing acne severity. MIDOO Smart Skin Imager was used to capture UVAF and analyze the correlation between fluorescence colors and acne lesions. Techniques such as two-photon excited fluorescence microscopy, scanning electron microscopy, western blot, and high performance liquid chromatography-mass spectrometry (HPLC-MS/MS) were used to examine the biochemical composition and structure of comedonal plugs and follicular casts associated with different fluorescence colors. We found that green fluorescence correlates with non-inflammatory acne lesions (comedones), while orange-red fluorescence shows no correlation with either type of lesion. Green fluorescence is associated with higher levels of keratin, indicating keratinization, while orange-red fluorescence is associated with porphyrin from S. epidermidis. UVAF color differences - orange-red are from porphyrins and green from keratin. This distinction helps to understand the structural and physiological bases of facial fluorescence, with potential implications for clinical evaluations of acne.

KPTr: Key point transformer for LiDAR-based 3D object detection

Accurate 3D environmental perception is crucial for ensuring the safety of autonomous vehicles. However, many existing point-based 3D object detection methods rely only on independent point features to extract spatial information and ignore the dependencies between points. Meanwhile, these methods usually use max-pooling operations to aggregate neighboring point features without considering the stability of local geometric information. To address these issues, this paper proposes a novel two-stage LiDAR-based key point transformer that consists of a region proposal network module and a region-based transformer module to enhance 3D detection performance. Firstly, unprocessed raw point clouds are taken as the input of the model to preserve detailed spatial structural information. Secondly, in the region proposal network an enhanced PointNet++ is used to segment foreground points and generate high-quality first-stage proposals. Then, the representation of proposals is improved by the region-based transformer module, which constructs long-range key point cloud relationship features within the regions of interest and diffuses spatial information into the feature channel layers. Additionally experiments on the KITTI dataset show that detection performance of the proposed method surpasses most mainstream 3D object detection methods and achieves a detection speed of 14.28 Hz, striking a good balance between detection accuracy and speed. Finally, the proposed method is deployed on real vehicle hardware platforms for online detection, thereby proving its industrial value.

TCRcost: a deep learning model utilizing TCR 3D structure for enhanced of TCR-peptide binding.

Predicting TCR-peptide binding is a complex and significant computational problem in systems immunology. During the past decade, a series of computational methods have been developed for better predicting TCR-peptide binding from amino acid sequences. However, the performance of sequence-based methods appears to have hit a bottleneck. Considering the 3D structures of TCR-peptide complexes, which provide much more information, could potentially lead to better prediction outcomes. In this study, we developed TCRcost, a deep learning method, to predict TCR-peptide binding by incorporating 3D structures. TCRcost overcomes two significant challenges: acquiring a sufficient number of high-quality TCR-peptide structures and effectively extracting information from these structures for binding prediction. TCRcost corrects TCR 3D structures generated by protein structure tools, significantly extending the available datasets. The main and side chains of a TCR structure are separately corrected using a long short-term memory (LSTM) model. This approach prevents interference between the chains and accurately extracts interactions among both adjacent and global atoms. A 3D convolutional neural network (CNN) is designed to extract the atomic features relevant to TCR-peptide binding. The spatial features extracted by the 3DCNN are then processed through a fully connected layer to estimate the probability of TCR-peptide binding. Test results demonstrated that predicting TCR-peptide binding from 3D TCR structures is both efficient and highly accurate with an average accuracy of 0.974 on precise structures. Furthermore, the average accuracy on corrected structures was 0.762, significantly higher than the average accuracy of 0.375 on uncorrected original structures. Additionally, the average root mean square distance (RMSD) to precise structures was significantly reduced from 12.753Å for predicted structures to 8.785Å for corrected structures. Thus, utilizing structural information of TCR-peptide complexes is a promising approach to improve the accuracy of binding predictions.

Vul-LMGNNs: Fusing language models and online-distilled graph neural networks for code vulnerability detection

Code Language Models (codeLMs) and Graph Neural Networks (GNNs) are widely used in code vulnerability detection. However, a critical yet often overlooked issue is that GNNs primarily rely on aggregating information from adjacent nodes, limiting structural information transfer to single-layer updates. In code graphs, nodes and relationships typically require cross-layer information propagation to fully capture complex program logic and potential vulnerability patterns. Furthermore, while some studies utilize codeLMs to supplement GNNs with code semantic information, existing integration methods have not fully explored the potential of their collaborative effects.To address these challenges, we introduce Vul-LMGNNs that integrates pre-trained CodeLMs with GNNs, leveraging knowledge distillation to facilitate cross-layer propagation of both code semantic knowledge and structural information. Specifically, Vul-LMGNNs utilizes Code Property Graphs (CPGs) to incorporate code syntax, control flow, and data dependencies, while employing gated GNNs to extract structural information in the CPG. To achieve cross-layer information transmission, we implement an online knowledge distillation (KD) program that enables a single student GNN to acquire structural information extracted from a simultaneously trained counterpart through an alternating training procedure. Additionally, we leverage pre-trained CodeLMs to extract semantic features from code sequences. Finally, we propose an ”implicit-explicit” joint training framework to better leverage the strengths of both CodeLMs and GNNs. In the implicit phase, we utilize CodeLMs to initialize the node embeddings of each student GNN. Through online knowledge distillation, we facilitate the propagation of both code semantics and structural information across layers. In the explicit phase, we perform linear interpolation between the CodeLM and the distilled GNN to learn a late fusion model. The proposed method, evaluated across four real-world vulnerability datasets, demonstrated superior performance compared to 17 state-of-the-art approaches. Our source code can be accessed via GitHub: https://github.com/Vul-LMGNN/vul-LMGGNN.

INCOMPLETE multi-view clustering based on low-rank adaptive graph learning

The challenge of acquiring complete data has led to substantial progress in incomplete multi-view clustering (IMVC) methods. Because graph structures can be excellent representations of data structure relationships, exceptional performance in handling incomplete data is demonstrated by graph-based methods at present. However, these methods still have their limitations. Most incomplete multi-view algorithms primarily focus on local information, neglecting global information. Therefore, these methods cannot dynamically recover the structural relationships in incomplete data by harnessing potential information from multiple perspectives and overall structural information. In response to the aforementioned concerns, we introduced an IMVC based on low-rank adaptive graph learning (IMVC-LAGL). This method initially constructs an affinity matrix based on the inter-view adjacency relationships. It also utilizes tensor low-rank constraints and consensus representation learning to explore higher-order correlations among different views. Subsequently, it adaptively reconstructs the incomplete graph structure to ultimately obtain a complete affinity relationship. It leads to excellent clustering results by integrating relevant information within views, overall structural information and potential information from multiple perspectives. We conducted experiments comparing our algorithm with eight incomplete multi-view algorithms using five different evaluation metrics. The results show that our algorithm achieves the best clustering results across eight datasets with varying missing rates. Particularly in the BBCSport dataset and YaleB dataset, the clustering accuracy of our algorithm is improved by 19.83 % and 16.41 %, respectively, compared with the second-best algorithm, under a 50 % missing rate.

Provenance-based APT campaigns detection via masked graph representation learning

Advanced Persistent Threats (APTs) are well-planned, persistent, and highly stealthy cyberattacks designed to steal confidential information or disrupt specific target systems. Recent studies have used system audit logs to construct provenance graphs that describe system interactions to detect potentially malicious activities. Although they are effective, they still suffer from problems such as the need for a priori knowledge, lack of attack data, and high computational overhead that limit their application. In this paper, we propose a self-supervised learning-based APT detection model, APT-MGL, which learns the embedded representations of nodes through a graph mask self-encoder and transforms the detection problem into an outlier detection problem for malicious nodes. APT-MGL characterizes the behavior of nodes based on node type, action, and interaction frequency, and fuses the features through a multi-head self-attention mechanism. Then the node embedding is obtained by combining graph features and structural information using masked graph representation learning. Finally, the unsupervised outlier detection method is used to analyze the computed embeddings and obtain the final detection results. The experimental results show that APT-MGL outperforms existing monitoring models and achieves a small overhead.

Multi-level discriminator based contrastive learning for multiplex networks

Graph embedding is a technique for obtaining low-dimensional representations of nodes across diverse networks, which may then be used for various downstream tasks and applications. When it applies to heterogeneous networks, it is hard to handle heterogeneous networks because they usually contain different types of nodes and edges with more semantic and structural information. Recently, contrastive learning has developed as the preferred strategy for dealing with unsupervised heterogeneous graph embedding to reduce the cost of human label annotation. However, most multi-view contrastive learning approaches calculate the model’s loss only based on the mutual dependence between the node representation and graph representation. These approaches ignore that both node attributes and node clustering contain discriminative content. To solve this issue, we propose a model called Multi-Level Discriminator-based Contrastive Learning for Multiplex Networks (MLDCL). This model adopts a multi-level multi-discriminator-based approach that can simultaneously learn the global-level structural information, node-level attribute information, and local-level clustering information. Moreover, an augmentation strategy in the contrast learning process from the spectral domain is proposed to improve the representation and discriminative ability of MLDCL. Numerous tests with node clustering and classification tasks on widely used datasets demonstrate the efficacy of the proposed approach.

Knowledge Representation Learning Method Based on Semantic Enhancement of External Information

Background: Knowledge representation learning aims at mapping entity and relational data in knowledge graphs to a low-dimensional space in the form of vectors. The existing work has mainly focused on structured information representation of triples or introducing only one additional kind of information, which has large limitations and reduces the representation efficiency. Objective: This study aims to combine entity description information and textual relationship description information with triadic structure information, and then use the linear mapping method to linearly transform the structure vector and text vector to obtain the joint representation vector. Methods: A knowledge representation learning (DRKRL) model that fuses external information for semantic enhancement is proposed, which combines entity descriptions and textual relations with a triadic structure. For entity descriptions, a vector representation is performed using a bi-directional long- and short-term memory network (Bi-LSTM) model and an attention mechanism. For the textual relations, a convolutional neural network is used to vectorially encode the relations between entities, and then an attention mechanism is used to obtain valuable information as complementary information to the triad. Results: Link prediction and triadic group classification experiments were conducted on the FB15K, FB15K-237, WN18, WN18RR, and NELL-995 datasets. Theoretical analysis and experimental results show that the DRKRL model proposed in this paper has higher accuracy and efficiency compared with existing models. Conclusion: Combining entity description information and textual relationship description information with triadic structure information can make the model have better performance and effectively improve the knowledge representation learning ability.

Structure-guided discovery of bile acid derivatives for treating liver diseases without causing itch.

Chronic itch is a debilitating symptom profoundly impacting the quality of life in patients with liver diseases like cholestasis. Activation of the human G-protein coupled receptor, MRGPRX4 (hX4), by bile acids (BAs) is implicated in promoting cholestasis itch. However, the detailed underlying mechanisms remain elusive. Here, we identified 3-sulfated BAs that are elevated in cholestatic patients with itch symptoms. We solved the cryo-EM structure of hX4-Gq in a complex with 3-phosphated deoxycholic acid (DCA-3P), a mimic of the endogenous 3-sulfated deoxycholic acid (DCA-3S). This structure revealed an unprecedented ligand-binding pocket in MRGPR family proteins, highlighting the crucial role of the 3-hydroxyl (3-OH) group on BAs in activating hX4. Guided by this structural information, we designed and developed compound 7 (C7), a BA derivative lacking the 3-OH. Notably, C7 effectively alleviates hepatic injury and fibrosis in liver disease models while significantly mitigating the itch side effects.

MHW-GAN: Multidiscriminator Hierarchical Wavelet Generative Adversarial Network for Multimodal Image Fusion.

Image fusion technology aims to obtain a comprehensive image containing a specific target or detailed information by fusing data of different modalities. However, many deep learning-based algorithms consider edge texture information through loss functions instead of specifically constructing network modules. The influence of the middle layer features is ignored, which leads to the loss of detailed information between layers. In this article, we propose a multidiscriminator hierarchical wavelet generative adversarial network (MHW-GAN) for multimodal image fusion. First, we construct a hierarchical wavelet fusion (HWF) module as the generator of MHW-GAN to fuse feature information at different levels and scales, which avoids information loss in the middle layers of different modalities. Second, we design an edge perception module (EPM) to integrate edge information from different modalities to avoid the loss of edge information. Third, we leverage the adversarial learning relationship between the generator and three discriminators for constraining the generation of fusion images. The generator aims to generate a fusion image to fool the three discriminators, while the three discriminators aim to distinguish the fusion image and edge fusion image from two source images and the joint edge image, respectively. The final fusion image contains both intensity information and structure information via adversarial learning. Experiments on public and self-collected four types of multimodal image datasets show that the proposed algorithm is superior to the previous algorithms in terms of both subjective and objective evaluation.

Synchrotron-aided exploration of REE recovery from coal fly ashes within a Canadian context

Coal ashes in Canada have gained attention as a potential source for recovering rare earth elements (REE) from industrial waste. However, the complex chemical properties of coal ashes have made it difficult to determine the desirability, feasibility, and viability of REE recovery. To address this issue, this study systematically investigated distribution and structural information, speciation and chemical-binding state, and purity and extraction capacity of REE in multiple Canadian coal ashes (i.e., 2 fly ash and 1 bottom ash samples) through synchrotron-based X-ray fluorescence mapping and adsorption spectrum analyses, as well as high-resolution REE sequential extraction quantitation. The results showed that Y, Ce, and La were present in the glass phase of the bottom ash, and the distributions of these REE elements correlated with Ca. The XANES analysis revealed that the dominant form of REE in coal fly ash (CFA) was REE oxides, indicating a transformation during combustion, while Y2O3 and Y2(CO3)3 were the predominant Y species identified in CFA. The study found that there is no correlation between P and REEs, suggesting that REEs in CFA may exist as discrete particles rather than being associated with amorphous glass. The extractability of REEs in bottom ash samples was lower than that in fly ash samples. Additionally, the benefits of REE recovery were estimated to be USD 99.82 to 215.21 per ton of fly ash through life cycle analysis, indicating that REE recovery from fly ashes is a promising path to supplement the REE supply chain in Canada.

Retinal fundus image super-resolution based on generative adversarial network guided with vascular structure prior

Many ophthalmic and systemic diseases can be screened by analyzing retinal fundus images. The clarity and resolution of retinal fundus images directly determine the effectiveness of clinical diagnosis. Deep learning methods based on generative adversarial networks are used in various research fields due to their powerful generative capabilities, especially image super-resolution. Although Real-ESRGAN is a recently proposed method that excels in processing real-world degraded images, it suffers from structural distortions when super-resolving retinal fundus images are rich in structural information. To address this shortcoming, we first process the input image using a pre-trained U-Net model to obtain a structural segmentation map of the retinal vessels and use the segmentation map as the structural prior. The spatial feature transform layer is then used to better integrate the structural prior into the generation process of the generator. In addition, we introduce channel and spatial attention modules into the skip connections of the discriminator to emphasize meaningful features and accordingly enhance the discriminative power of the discriminator. Based on the original loss functions, we introduce the L1 loss function to measure the pixel-level differences between the segmentation maps of retinal vascular structures in the high-resolution images and the super-resolution images to further constrain the super-resolution images. Simulation results on retinal image datasets show that our improved algorithm results have a better visual performance by suppressing structural distortions in the super-resolution images.

Structural characterization, antioxidant activities and physicochemical properties analysis of a galactose-rich exopolysaccharide produced by Limosilactobacillus fermentum YL-11

Lactobacilli-derived exopolysaccharides (EPSs) are widely regarded as safe and have been employed as potential natural, non-toxic additives and functional agents in food industry. The structural properties of a galactose-rich exopolysaccharide produced by Limosilactobacillus fermentum YL-11 (YL-11 EPS) were investigated with chemical and instrumental analysis. The YL-11 EPS was determined to have a molecular weight (Mw) of 13.2 kDa. Nine different types of glycosidic bonds were identified in YL-11 EPS. The inferred structure of the YL-11 EPS contained the backbone of →2)-β-D-Galp-(1→[3)-β-D-Galp-(1]4 → 3)-β-D-Galp-(1 → 3)-β-D-Galp-(1→, with two side chains, β-D-Galp-(1 → 6)-α-D-Manp-(1 → 6)-[α-D-Glcp-1→2]-α-D-Manp-(1 → 3)-[α-D-Glcp-1→6]-α-D-Glcp-(1 → 3)-[α-D-Glcp-1→6]-α-D-Glcp-(1→ with an acetyl group at the O-3 position of →2,6-α-D-Manp-1→ and α-L-Araf-(1→, which were present at the C6 location of the 1,3-β-D-Galp. The YL-11 EPS showed concentration-dependent antioxidant activities and the DPPH, ABTS, hydroxyl groups and superoxide radicals scavenging rates of 10 mg/mL YL-11 EPS reached 61%, 74%, 35.4% and 40%, respectively. The YL-11 EPS also has a beneficial protective impact on H2O2-injured cells, which pretreatment with YL-11 EPS can substantially decrease ROS levels and enhance SOD enzyme activity. Additionally, the YL-11 EPS exhibited good thickening properties and thermal stability. The EPS's structural information and physicochemical properties provide a basis for studying their structure-function relationships.