Computational Biology and Chemistry最新文献

{"title":"QDCRNet: Quantum dilated convolutional recurrent network for virus detection using gene expression data","authors":"S. Karthi ,&nbsp;T. Ramalingam ,&nbsp;R. Iyswarya ,&nbsp;D. Arul Kumar","doi":"10.1016/j.compbiolchem.2025.108625","DOIUrl":"10.1016/j.compbiolchem.2025.108625","url":null,"abstract":"<div><div>Viral Infections cause several common human illnesses, like the common cold, flu, mouth blisters, and chickenpox. However, numerous viruses, including rabies, hepatitis, Ebola, avian flu, and coronavirus cause significant health risks due to their high transmission rates. Infections are obligate intracellular parasites that depend on host cellular organisms, resources, and replication for their reproduction and spread. Timely and precise identification of these viruses is vital for appropriate treatment and preventing further spread. However common challenges such as, non-specific symptoms, variability in viral expression, and delayed testing often complicate timely diagnosis. To address this issue, a powerful module named Quantum Dilated Convolutional Recurrent Network (QDCRNet) has been developed for virus detection. Firstly, gene expression data is given into data transformation, and it is done by the Box-Cox transformation. Then, Feature Selection (FS) is performed using Gower distance and mutual information to select the virus-affected region. Finally, detection of the virus is done using QDCRNet, which is the integration of Quantum Dilated Convolutional Neural Network (QDCNN) and Deep Recurrent Neural Network (DRNN) model. The proposed QDCRNet has achieved a great performance with an accuracy of 90.80 %, sensitivity of 90.50 % and specificity of 90.40 %.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"120 ","pages":"Article 108625"},"PeriodicalIF":3.1,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144880037","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"METACONE: A scalable framework for exploring the conversion cone of metabolic networks","authors":"Álvaro Altamirano ,&nbsp;Ignacio Tapia ,&nbsp;Vicente Acuña ,&nbsp;Daniel Garrido ,&nbsp;Pedro A. Saa","doi":"10.1016/j.compbiolchem.2025.108607","DOIUrl":"10.1016/j.compbiolchem.2025.108607","url":null,"abstract":"<div><div>Elementary Conversion Modes (ECMs) – a subset of Elementary Flux Modes (EFMs) – capture the entire production/consumption potential of a metabolic network, providing a more practical view of its interactions with the environment. Despite its reduced size, the set of ECMs is too large for exhaustive enumeration in models reaching genome scale. To address this limitation, we have developed METACONE (METAbolic Conversion cOne for Network Exploration), a scalable algorithm for the computation of a representative linear basis of the conversion cone, the subspace in which all ECMs lie. Two METACONE variants are proposed based on the solution of a series of linear programs following different heuristics. We evaluated the performance of the variants on metabolic models of different sizes, demonstrating their scalability. We further analyzed the resulting basis to explore metabolic capabilities under different environmental conditions in <em>Escherichia coli</em>, identifying metabolic patterns consistent with reported data. Finally, we applied the algorithm to explore metabolic interactions in a microbial consortium of <em>Phocaeicola dorei</em> and <em>Lachnoclostridium symbiosum</em>, recapitulating known cross-feeding interactions and suggesting new possibilities. We envision METACONE as a valuable tool for understanding microbial metabolism in increasingly complex consortia while addressing current scalability challenges.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"120 ","pages":"Article 108607"},"PeriodicalIF":3.1,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144810159","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"FEAOF: A transferable framework applied to prediction of hERG-related cardiotoxicity","authors":"Bowen Zhao ,&nbsp;Zhenghui Chang ,&nbsp;Mengqi Huo","doi":"10.1016/j.compbiolchem.2025.108622","DOIUrl":"10.1016/j.compbiolchem.2025.108622","url":null,"abstract":"<div><div>Inhibition of the hERG (human ether-a-go-go-related gene) channel by drug molecules can lead to severe cardiac toxicity, resulting in the withdrawal of many approved drugs from the market or halting their development in later stages. These findings highlight the pressing need to evaluate hERG blockade during drug development. We propose a novel framework for feature extraction and aggregation optimization (FEAOF), which primarily consists of a feature extraction module and an aggregation optimization module. The model integrates diverse ligand representations, including molecular fingerprints, descriptors, and graphs, as well as ligand–receptor interaction features. Based on this integration, we further optimize the algorithmic framework to achieve precise predictions of compounds cardiac toxicity. Two independent test sets exhibiting pronounced structural dissimilarity from the training data were constructed to rigorously assess the model’s generalization ability. The results demonstrate that the FEAOF model exhibits strong robustness compared to seven baseline models, achieving F1 score of 66.1 % and 68.1 %. Notably, when benchmarked against five existing models on two external test sets, FEAOF also achieved the highest or near-highest scores across all key evaluation metrics. Importantly, this model can be easily adapted for other drug-target interaction prediction tasks. It is made available as open source under the permissive MIT license at <span><span>https://github.com/ConfusedAnt/FEAOF</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"120 ","pages":"Article 108622"},"PeriodicalIF":3.1,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144810229","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Regulation of oxytocin receptor by zinc coumarin derivatives: a mechanistic approach to alleviate anxiety and enhance folliculogenesis in letrozole-induced PCOS in zebrafish model","authors":"Karthikeyan Ramamurthy ,&nbsp;Sanjay Gopi ,&nbsp;S. Madesh ,&nbsp;B. Aswinanand ,&nbsp;Preeya Negi ,&nbsp;Jubie Selvaraj ,&nbsp;Mikhlid H. Almutairi ,&nbsp;Bader O. Almutairi ,&nbsp;S. Karthick Raja Namasivayam ,&nbsp;Kathiravan Muthu Kumaradoss ,&nbsp;Jesu Arockiaraj","doi":"10.1016/j.compbiolchem.2025.108610","DOIUrl":"10.1016/j.compbiolchem.2025.108610","url":null,"abstract":"<div><div>Polycystic ovary syndrome (PCOS) is a common endocrine disorder in women, characterized by insulin resistance and mood disturbances. The therapeutic potential of traditional treatments is often limited by side effects, highlighting the need for novel interventions. This study investigated the efficacy of newly synthesized thiosemicarbazone coumarin zinc complexes, TSCO6-Zn (T1) and TSCO13-Zn (T2) were assessed in a letrozole-induced PCOS <em>in-vivo</em> zebrafish model. Synthesis involved Pechmann condensation to form 7-hydroxy-4-methylcoumarin, followed by coordination with zinc. The compounds’ targets were predicted via BindingDB, with molecular docking confirming interactions with PCOS-related proteins. <em>In vivo</em> toxicity was assessed in zebrafish embryos exposed to T1 and T2 (up to 150 µM), examining behavioral assays, body weight, lipid profile, GSI (%) and folliculogenesis. In addition to that, HPLC testosterone quantification and qPCR for gene expression analysis were employed for <em>20β-hsd</em>, <em>cyp19a1a</em>, <em>dennd1a</em>, <em>tox3, oxtr</em>, <em>mTOR, pik3cd</em>, and <em>drd2a</em>. T1 and T2 markedly reduced anxiety, lowered testosterone, and enhanced follicular maturation, with no toxicity observed up to 50 µM. Docking studies demonstrated a high affinity of T1 and T2 for key metabolic and neurobehavioral targets such as <em>drd2a, oxtr, mTOR,</em> and <em>pik3cd</em>. Significant improvements were noted in body weight, lipid profiles, oxidative stress markers, and normalized gene expressions involved in steroidogenesis and metabolic pathways in letrozole-induced PCOS in the zebrafish. T1 and T2 effectively mitigate metabolic and neurobehavioral disturbances associated with PCOS in the zebrafish model, suggesting their potential as comprehensive therapeutic agents. Their multi-target approach could provide a basis for advanced PCOS treatment strategies.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"119 ","pages":"Article 108610"},"PeriodicalIF":3.1,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144779858","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Harnessing transcriptome data of Viola inconspicua for discovery of novel cyclotides and AEP ligases","authors":"S. Hamsa,&nbsp;Suhani Pathak,&nbsp;Reema Mishra,&nbsp;Aparajita Mohanty","doi":"10.1016/j.compbiolchem.2025.108614","DOIUrl":"10.1016/j.compbiolchem.2025.108614","url":null,"abstract":"<div><div>Cyclotides are plant-derived cyclic peptides with three conserved disulfide linkages, forming a cyclic cystine knot (CCK) motif. This CCK topology makes them ultra stable structures and resistant to thermal and chemical degradation. Cyclotides are known to exhibit, anti-HIV, antimicrobial, cytotoxic, hemolytic and pesticidal bioactivities. They have been reported in six angiosperm families (Cucurbitaceae, Fabaceae, Poaceae, Rubiaceae, Solanaceae, Violaceae). The identification of novel cyclotides is the first important step for investigating their potential applications in agriculture and therapeutics. To address this need, the present study employed a de novo transcriptome assembly of <em>Viola inconspicua</em> root and shoot tissues. HMM-based searches were used to identify novel cyclotides and the enzymes involved in their biosynthesis, specifically asparaginyl endopeptidases (AEPs). The analysis revealed six types of cyclotide precursor (CP) gene architectures and 68 novel cyclotides of which 31 and 19 were exclusive to roots and shoots respectively, and 18 were common in both tissues. Assessment of potential bioactivity of 68 novel cyclotides was investigated by analysing their physicochemical characteristics, loop sequence variations and phylogenetic studies. Furthermore, the analysis revealed 40 AEP isoforms. Two of these were identified as potential peptide asparaginyl ligases (PALs), important for the cyclization of cyclotides. Moreover, comparison of homology models of potential PALs with experimentally validated structure of PAL from <em>Viola yedoensis</em> (VyPAL2) revealed high structural homology. In summary, this study reveals tissue specific diversity of cyclotides in <em>V. inconspicua</em>; identifies novel cyclotides, provides insights on CP gene architectures and structure of potential PALs.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"119 ","pages":"Article 108614"},"PeriodicalIF":3.1,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144773002","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Identification of hub genes as potential diagnostic biomarkers for cervical cancer: A bioinformatic approach","authors":"Tara Chand ,&nbsp;Pankaj Vaishanavaa ,&nbsp;Ashwini Kumar Dubey ,&nbsp;Gauri Misra","doi":"10.1016/j.compbiolchem.2025.108605","DOIUrl":"10.1016/j.compbiolchem.2025.108605","url":null,"abstract":"<div><h3>Background</h3><div>Cervical cancer remains a prevalent malignancy with rising incidence, primarily due to sexual transmission, persistent HPV infection, and delayed screening. Identifying new biomarkers is critical for improved diagnosis, prognosis, and treatment of cervical cancer. This study utilized integrated bioinformatics to identify potential biomarkers by analysing gene expression data from the GEO database.</div></div><div><h3>Methods</h3><div>Four GEO microarray datasets (GSE7410, GSE7803, GSE52903, GSE67522) were analysed using GEO2R to identify DEGs with an adjusted p-value &lt;0.05. Common DEGs were visualized using Venn diagrams. Protein-protein interaction network was constructed using STRING to identify hub genes. Gene Ontology (GO) and KEGG pathway analyses were performed to investigate biological functions and pathways. The Human Protein Atlas (HPA) was used for <em>in silico</em> validation of protein expression via immunohistochemistry. Kaplan-Meier survival analysis was performed to determine the prognostic value of hub genes.</div></div><div><h3>Results</h3><div>Analysis revealed 684 common DEGs across the datasets (446 upregulated, 238 downregulated). The top 20 upregulated DEGs from GSE67522 were used for heatmap construction and PPI analysis, leading to the identification of nine key hub genes. GO and KEGG analyses showed that six of these were significantly involved in cell cycle regulation and tumorigenic pathways. These hub genes were validated for their protein expression through HPA data.</div></div><div><h3>Conclusion</h3><div>Six hub genes (CCNB2, AURKA, CDC20, CDT1, CENPF, and KIF2C) were identified as potential biomarkers for cervical cancer management.</div></div><div><h3>Impact</h3><div>These findings provide valuable insight into the molecular profiles of genes that play significant roles in cervical cancer for translational outcomes in diagnosis.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"119 ","pages":"Article 108605"},"PeriodicalIF":3.1,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144773003","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A unified graph-based approach for protein function prediction using AlphaFold structures and sequence features","authors":"Thi-Tuyen Nguyen ,&nbsp;Wenqing Zheng ,&nbsp;Van-Nui Nguyen ,&nbsp;Nguyen Quoc Khanh Le ,&nbsp;Matthew Chin Heng Chua","doi":"10.1016/j.compbiolchem.2025.108609","DOIUrl":"10.1016/j.compbiolchem.2025.108609","url":null,"abstract":"<div><div>Predicting protein function is a key challenge in computational biology with broad implications for understanding biological systems and disease mechanisms. Traditional deep learning approaches rely heavily on protein sequence data and protein–protein interaction (PPI) networks, often neglecting structural information due to limited availability of experimentally resolved protein structures. The advent of AlphaFold, which predicts protein structures with near-atomic accuracy, provides an opportunity to integrate structural context into function prediction. In this study, we propose StructSeq2GO, a novel hybrid model that combines structural and sequence information. StructSeq2GO employs graph representation learning to extract structural features from AlphaFold-predicted protein structures and integrates them with sequence embeddings derived from the ProteinBERT language model to predict Gene Ontology (GO) labels. Experimental evaluations demonstrate that StructSeq2GO achieves state-of-the-art performance across three GO domains, with <span><math><msub><mrow><mi>F</mi></mrow><mrow><mi>m</mi><mi>a</mi><mi>x</mi></mrow></msub></math></span> scores of 0.485, 0.681, and 0.663, AUC scores of 0.764, 0.939, and 0.891, and AUPR scores of 0.688, 0.763, and 0.702 for the Biological Process (BPO), Cellular Component (CCO), and Molecular Function (MFO) ontologies, respectively. These results highlight the critical importance of structural information and the efficacy of ProteinBERT in enhancing protein function prediction, as structure provides spatial and biochemical context not captured by sequence alone. The model’s performance is influenced by the quality of AlphaFold structural predictions and may benefit from future improvements in structure confidence modeling. Additionally, extending StructSeq2GO to predict pathway-level or disease-related annotations could broaden its biological utility.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"120 ","pages":"Article 108609"},"PeriodicalIF":3.1,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144841014","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimization enabled ResNet features with transfer learning for Alzheimer’s disease detection","authors":"Deepthi K. Moorthy ,&nbsp;P. Chinnasamy ,&nbsp;P. Nagaraj","doi":"10.1016/j.compbiolchem.2025.108613","DOIUrl":"10.1016/j.compbiolchem.2025.108613","url":null,"abstract":"<div><div>Millions of individuals worldwide suffer from Alzheimer's Disease (AD), a debilitating degenerative condition. Early detection of Alzheimer's disease is critical to ensure effective treatment and better patient outcomes. In the past few years, advanced medical imaging techniques, particularly MRI, have shown potential for diagnosing Alzheimer's disease. However, developing accurate and efficient techniques for Alzheimer's disease detection offcuts a demanding duty suitable to the complication of medical images and the limited availability of labelled data. The early detection of Alzheimer's disease is critical for effective treatment and management of this debilitating neurodegenerative condition. The research proposes a novel method for Alzheimer's disease detection using an optimization-enabled ResNet feature extraction technique with transfer learning that is proposed by combining LeNet and VGG networks. The pre-processing was done using image resizing and median filter and the featureextraction was conducted using the proposed Walrus Optimization Algorithm-Residual neural network (WOA-ResNet), where WOA is employed for training ResNet. The conducted experiments with the Alzheimer’s dataset achieved a higher accuracy using the proposed LeNet-VGG method. The findings suggest that optimization-enabled ResNet feature extraction with LeNet-VGG networks can significantly improve the accuracy of Alzheimer's disease detection. The presented method achieved maximum accuracy value of 95.37 %, sensitivity value of 97.24 % and specificity value of 93.73 %.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"119 ","pages":"Article 108613"},"PeriodicalIF":3.1,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144862724","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel network for resolving subjective masking differences and accurate thyroid nodule diagnosis","authors":"Zhiyuan Ouyang ,&nbsp;Simei Huang ,&nbsp;Liuju Liang ,&nbsp;Jianing Xu ,&nbsp;Caifen Wei ,&nbsp;Yi Zhang ,&nbsp;Hancheng Jiang ,&nbsp;Haifeng Tang ,&nbsp;Lu Wang ,&nbsp;Lin Wang ,&nbsp;Xiangzhi Li ,&nbsp;Zhenbing Liu ,&nbsp;Ruojie Zhang ,&nbsp;Lian Qin ,&nbsp;Xiaobo Yang","doi":"10.1016/j.compbiolchem.2025.108572","DOIUrl":"10.1016/j.compbiolchem.2025.108572","url":null,"abstract":"<div><h3>Background:</h3><div>Over the past three decades, there has been a significant increase in the incidence of thyroid cancer. Ultrasound serves as a non-invasive tool in differentiating between benign and malignant thyroid nodules. However, its reliance on manual input can often lead to subjective bias.</div></div><div><h3>Purpose:</h3><div>This study proposes a novel network architecture committed to diminishing subjective bias led by manual masks and enhancing the accuracy of the current models. It amalgamates multi-scale features for the effective classification of thyroid nodules.</div></div><div><h3>Methods:</h3><div>The innovative model, deemed APSNet, finds inspiration from active and passive systems. It incorporates attention mechanisms to augment nodule recognition. The model underwent training on a localized ultrasound image dataset and was tested using an external datasets TDID and TN3K. The assessment of its performance involved metrics such as Dice, IoU, F1, Acc, Sen, Spe, Ppv, Npv, and AUC, followed by statistical tests including the Friedman and DeLong tests.</div></div><div><h3>Results:</h3><div>APSNet outperformed existing models across multiple metrics, achieving an Acc of 0.9259, F1 score of 0.9540, and AUC of 0.9243 on the TDID dataset, and an Acc of 0.9287, F1 score of 0.9001, sensitivity of 0.9273, and AUC of 0.9290 on the TN3K dataset. The DeLong test confirmed its superiority, indicating statistically significant improvements over other models. Ablation Study confirms the effectiveness of Dual-System design and the potention of Transformer-based backbone.</div></div><div><h3>Conclusions:</h3><div>APSNet offers a remarkable stride forward in thyroid nodule diagnosis by effectively addressing subjectivity and amplifying feature extraction capabilities. It proffers a more accurate and dependable diagnostic tool to clinicians.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"120 ","pages":"Article 108572"},"PeriodicalIF":3.1,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144852412","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Systemic Lupus Erythematosus prediction using Epistatic-Quantile Fusion Transformer network with integrated multi-omics and clinical data","authors":"Manoj B. Chandak,&nbsp;Abhijeet R. Raipurkar,&nbsp;Sunita G. Rawat","doi":"10.1016/j.compbiolchem.2025.108617","DOIUrl":"10.1016/j.compbiolchem.2025.108617","url":null,"abstract":"<div><div>Systemic Lupus Erythematosus (SLE) is a complex autoimmune disorder with heterogeneous symptoms and overlapping clinical presentations, making early prediction extremely difficult. Traditional models often fail to integrate high-dimensional multi-omics data and EHR records effectively, primarily due to their inability to handle biological variability, data imbalance, and complex feature dependencies. To address these gaps, the study proposes Epistatic-Quantile Fusion Transformer (EQF-T), a unified framework that introduces multiple novel components. Initially, for pre-processing, the Beta-Variational Rank-ordered Quantile Autoencoder (Beta-VARQA) is used, which combines Beta-divergence, Rank-ordered Quantile Filtering, and Variational Autoencoding to denoise and normalize heterogeneous inputs, retaining biologically significant patterns. For feature extraction, the framework incorporates Epistatic Attention fused Multi-Omics Laplacian Transformer (EA-MLT), which captures intricate dependencies and Epistatic Synergistic effects, essential for understanding the dynamic progression of SLE. This EA-MLT employs Epistatic Attention to capture higher-order gene-gene interactions and integrates the Multi-Omics Laplacian Transformer (MOLT), which uses a Laplacian Attention Mechanism to model structural dependencies across omics layers. The final classification is performed by SLE-Net (SLE Prediction Network), an end-to-end deep learning model designed to analyze fused data and provide interpretable outputs. Together, these components enable EQF-T to effectively learn from complex, high-dimensional biological and clinical data. Further, the proposed model achieves superior performance with 99.82 % accuracy, 99.78 % precision, 99.76 % recall, 99.77 % F1-score, and 99.8 % ROC-AUC, demonstrating its reliability and potential for precise SLE prediction.</div></div>","PeriodicalId":10616,"journal":{"name":"Computational Biology and Chemistry","volume":"120 ","pages":"Article 108617"},"PeriodicalIF":3.1,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144852411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}