Computers in biology and medicine最新文献

{"title":"Interactive multi-hypergraph inferring and channel-enhanced and attribute-enhanced learning for drug-related side effect prediction","authors":"Ping Xuan ,&nbsp;Shien Wu ,&nbsp;Hui Cui ,&nbsp;Peiru Li ,&nbsp;Toshiya Nakaguchi ,&nbsp;Tiangang Zhang","doi":"10.1016/j.compbiomed.2024.109321","DOIUrl":"10.1016/j.compbiomed.2024.109321","url":null,"abstract":"<div><div>Identifying the potential side effects for the interested drugs can help reduce harm to patients caused by drugs in clinical use and decrease the risk of drug development failure. Multiple functionally similar drugs often have multiple similar side effects, resulting in the closed relationships among these nodes. However, most of previous methods did not completely encode the features from the biological perspective to mine the complex associations between the drugs and side effects. A prediction model based on interactive multi-hypergraph inferring and channel-enhanced and attribute-enhanced learning, ICAL, was proposed to fuse the global correlations reflected by multiple hypergraphs and to learn the attributes of a pair of drug and side effect nodes enhanced by the channels and attributes. First, we designed a hypergraph architecture where a hyperedge reflects the complex correlations between a single drug (side effect) and all the drugs and side effects, and the entire hypergraph composed of the hyperedges reveals the global correlations of all the drugs and side effects. Two hypergraphs were established based on two types of drug similarities, and each hypergraph implies its specific complex relationships among multiple drugs and side effects. Second, we proposed an interactive hypergraph neural network to enable the learning of global correlation features of drugs and side effects from the two hypergraphs. It propagated the node features across multiple hypergraphs and encoded the context relationships within these hypergraphs. Besides, the attentions at the channel level and at the attribute level were proposed to integrate the semantic correlations among multiple channels and to encode the long-distance dependence within the attributes of a pair of drug and side effect. The experimental results based on cross-validation showed that our new model outperformed seven advanced prediction methods in terms of AUC, AUPR, and recall rates for the top-ranked candidates. The ablation studies showed the effectiveness of global correlation learning, node feature propagation across multiple hypergraphs, and channel and attribute enhanced pairwise attribute learning. The case studies on the candidate side effects related to five drugs further demonstrated ICAL’s effective application in discovering the reliable candidates.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109321"},"PeriodicalIF":7.0,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142615223","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Utilization of a hierarchical electrocardiogram classification model for enhanced biometric identification","authors":"YeJin Kim,&nbsp;Chang Choi","doi":"10.1016/j.compbiomed.2024.109254","DOIUrl":"10.1016/j.compbiomed.2024.109254","url":null,"abstract":"<div><div>Emerging research on artificial intelligence (AI) has leveraged the unique properties of electrocardiogram (ECG) signals for user identification. ECG signals, known for their resistance to forgery and tampering, offer security advantages. However, these signals fluctuate in response to physical and cognitive stress. Despite their security benefits, these dynamic characteristics present challenges for consistent user identification owing to their variable amplitudes and shapes. To address these problems, we propose a 2-stage user identification system that integrates ECG signals and status information. This system classifies the user’s ECG status and uses the feature values in a second model to improve dynamic feature learning ability. This allows identification with high accuracy even in various stress states of the user. This increases the real-life usability of the ECG user identification system. The effectiveness of the proposed method was confirmed through a performance evaluation using CSU-BIODB(Chosun University-BIO Database) and the public MIT-BIH(Massachusetts Institute of Technology - Beth Israel Hospital Arrhythmia Laboratory) ST Change database, with identification accuracies of 92.08% and 95.83%, and f1-scores of 0.9207 and 0.9369, respectively. Compared with existing single user identification models, our approach demonstrated accuracy improvements of 9.3% and 36.76% for each database. These findings underscore the potential of the new 2-stage model for enhancing the practicality of ECG-based user identification systems and provide a promising foundation for future research on deep learning signal processing.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109254"},"PeriodicalIF":7.0,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142615900","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel semi-supervised learning model based on pelvic radiographs for ankylosing spondylitis diagnosis reduces 90% of annotation cost","authors":"Hao Li ,&nbsp;Dong Yin ,&nbsp;Baichuan Li ,&nbsp;Chong Liu ,&nbsp;Chunxiang Xiong ,&nbsp;Qie Fan ,&nbsp;Shuyu Yao ,&nbsp;Wenwen Huang ,&nbsp;Wenhao Li ,&nbsp;Jingda Zhang ,&nbsp;Hongmian Li","doi":"10.1016/j.compbiomed.2024.109232","DOIUrl":"10.1016/j.compbiomed.2024.109232","url":null,"abstract":"<div><h3>Objective:</h3><div>Our study aims to develop a deep learning-based Ankylosing Spondylitis (AS) diagnostic model that achieves human expert-level performance using only a minimal amount of labeled samples for training, in regions with limited access to expert resources.</div></div><div><h3>Methods:</h3><div>Our semi-supervised diagnostic model for AS was developed using 5389 pelvic radiographs (PXRs) from a single medical center, collected from March 2014 to April 2022. The dataset was split into a training set and a validation set with an 8:2 ratio, allocating 431 labeled images and the remaining 3880 unlabeled images for semi-supervised learning. The model’s performance was evaluated on 982 PXRs from the same center, assessing metrics such as AUC, accuracy, precision, recall, and F1 scores. Interpretability analysis was performed using explainable algorithms to validate the model’s clinical applicability.</div></div><div><h3>Results:</h3><div>Our semi-supervised learning model achieved accuracy, recall, and precision values of 0.891, 0.865, and 0.859, respectively, using only 10% of labeled data from the entire training set, surpassing human expert performance. Extensive interpretability analysis demonstrated the reliability of our model’s predictions, making the deep neural network no longer a black box.</div></div><div><h3>Conclusion:</h3><div>This study marks the first application of semi-supervised learning to diagnose AS using PXRs, achieving a 90% reduction in manual annotation costs. The model showcases robust generalization on an independent test set and delivers reliable diagnostic performance, supported by comprehensive interpretability analysis. This innovative approach paves the way for training high-performance diagnostic models on large datasets with minimal labeled data, heralding a cost-effective future for medical imaging research in big data analytics.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109232"},"PeriodicalIF":7.0,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142616388","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Measurement of ureteral length: Comparison of deep learning-based method and other estimation methods on CT and KUB","authors":"Kexin Wang ,&nbsp;Zheng Zhao ,&nbsp;Yi Liu ,&nbsp;Rile Nai ,&nbsp;Changwei Yuan ,&nbsp;Pengsheng Wu ,&nbsp;Jialun Li ,&nbsp;Xiaodong Zhang ,&nbsp;He Wang","doi":"10.1016/j.compbiomed.2024.109374","DOIUrl":"10.1016/j.compbiomed.2024.109374","url":null,"abstract":"<div><h3>Background</h3><div>Accurate preoperative assessment of ureteral length is crucial for effective ureteral stenting.</div></div><div><h3>Purpose</h3><div>Utilize a deep learning approach to measure ureter length on CT urography (CTU) images and compare the obtained results with those derived from other estimation methods.</div></div><div><h3>Methods</h3><div>In a retrospective cohort (cohort A, n = 411), CTU images were collected and used to develop a 3D deep learning model for the segmentation of bilateral ureters. The centerline of the ureters was determined based on the segmentation, and the length of the ureters was automatically obtained (CTU_ai). Another cohort (cohort B, n = 220) was collected as the hold-out test for the model. All patients in cohort B had KUB, non-contrast enhanced CT (CT NoC), and CTU images. Cohort B utilized eight measurement methods, with one annotated by two radiologists serving as the reference standard (CTU_ref) and the remaining seven as the studied methods, including three measurement methods applied to CTU (CTU_ai, CTU_oblique, CTU_slice), two applied to CT NoC (CT_oblique, CT_slice), and two applied to KUB (KUB_short, KUB_long). The results of the seven studied methods were compared to those of the reference in cohort B.</div></div><div><h3>Results</h3><div>Among the 220 patients (96 females, 124 males), 437 ureters were measured for length (218 left, 219 right), with a median length of 24.7 (IQR 23.2–26.2) cm. No significant differences were observed between genders or laterality (both <em>P</em> &gt; 0.05). Moreover, there was no correlation between ureteral length and age (r = −0.027, <em>P</em> = 0.573). The ureteral length measured by CTU_ai was not significantly different from that measured by CTU_ref (<em>P</em> = 0.514), whereas the length measured by the other studied methods was significantly different from that measured by CTU_ref (all <em>P</em> &lt; 0.001). The ICC values with their 95 % confidence intervals (CIs) for the comparison between the reference standard (CTU_ref) and the other measurement methods: CTU_ai (ICC = 0.852, 95 % CI 0.825–0.876), CTU_oblique (ICC = 0.351, 95 % CI -0.083-0.689), CTU_slice (ICC = 0.269, 95 % CI -0.095-0.573), CTU_oblique_slice (ICC = 0.059, 95 % CI -0.032-0.218), CTU_slice (ICC = 0.049, 95 % CI -0.028-0.188), KUB_short (ICC = 0.151, 95 % CI 0.051–0.247), and KUB_long (ICC = 0.147, 95 % CI 0.034–0.253). For CTU_ai, in 89.0 % of the ureters, the ureteral length deviation was within 20 mm of the reference standard, which was the highest among all the studied methods (all <em>P</em> &lt; 0.001).</div></div><div><h3>Conclusion</h3><div>The deep learning model offers a reliable and accurate tool for ureteral length measurement on CTU images, which could enhance the effectiveness of ureteral stenting procedures. Its performance surpasses traditional measurement methods, making it a promising technology for integration into clinical practice.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109374"},"PeriodicalIF":7.0,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142615419","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improving the identification of cancer driver modules using deep features learned from multi-omics data","authors":"Yang Guo,&nbsp;Lingling Liu,&nbsp;Aofeng Lin","doi":"10.1016/j.compbiomed.2024.109322","DOIUrl":"10.1016/j.compbiomed.2024.109322","url":null,"abstract":"<div><div>Identifying the cancer driver modules or pathways is crucial to understanding the fundamental mechanisms of cancer occurrence and progression. The rapid abundance of cancer omics data provides unprecedented opportunities to study the driver modules in cancer, and many computational methods have been developed in recent years. However, most existing methods have limitations in considering different types of cancer omics data and cannot effectively learn informative omics features for integrated identification of driver modules. In this paper, we introduce a new integrated framework to accurately identify the cancer driver modules by integrating the protein-protein interaction network, transcriptional regulatory network, gene expression and mutation data in cancer. We first develop a series of methods to learn the deep features of functional connectivity between genes in each omics data and then construct an integrated gene functional coherence network. Furthermore, we present a two-step module mining method to efficiently identify the cancer driver modules from the integrated gene functional coherence network. Systematic experiments in three cancer types demonstrate that the proposed framework can obtain more significant driver modules than most existing methods, and some identified driver modules are associated with clinical survival phenotypes.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109322"},"PeriodicalIF":7.0,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142616419","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unveiling AI's role in papilledema diagnosis from fundus images: A systematic review with diagnostic test accuracy meta-analysis and comparison of human expert performance","authors":"Paweł Marek Łajczak ,&nbsp;Sebastian Sirek ,&nbsp;Dorota Wyględowska-Promieńska","doi":"10.1016/j.compbiomed.2024.109350","DOIUrl":"10.1016/j.compbiomed.2024.109350","url":null,"abstract":"<div><h3>Background</h3><div>Papilledema is a condition, which is characterized by optic disc swelling due to increased intracranial pressure. Diagnostic modalities include fundus camera and other ophthalmology imaging techniques. The Frisén scale is used to grade the severity of this condition. In this paper, we investigate the application of artificial intelligence (AI) for detecting and grading papilledema from fundus images.</div></div><div><h3>Method</h3><div>Following the PRISMA guidelines for systematic reviews, a search of five databases (PubMed, Scopus, Web of Science, Embase, Cochrane) was conducted using MeSH terms related to AI and papilledema. The inclusion criteria were original articles that discussed AI applications for detecting or grading papilledema from fundus images. Extracted data included sensitivity, specificity, accuracy, and technical and demographic characteristics.</div></div><div><h3>Results</h3><div>The systematic review included 21 studies. In the meta-analysis, the pooled sensitivity and specificity were 0.97 and 0.98, respectively. High heterogeneity was observed (I² &gt; 96%). Deep learning models outperformed traditional machine learning algorithms, with detection models being more effective than grading models. Publication bias was observed with Deek's plot. Several publications compared AI to human experts, showing superiority or non-inferiority of computer algorithms to humans.</div></div><div><h3>Conclusions</h3><div>AI models show high diagnostic accuracy in detecting papilledema, often surpassing human experts in sensitivity, though not always in specificity. Despite limitations related to patient selection, image sourcing, and heterogeneity, AI holds potential to significantly improve diagnostic accuracy and clinical workflows in ophthalmology.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109350"},"PeriodicalIF":7.0,"publicationDate":"2024-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142603411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Biomechanical stress analysis of Type-A aortic dissection at pre-dissection, post-dissection, and post-repair states","authors":"Christina Sun ,&nbsp;Tongran Qin ,&nbsp;Asanish Kalyanasundaram ,&nbsp;John Elefteriades ,&nbsp;Wei Sun ,&nbsp;Liang Liang","doi":"10.1016/j.compbiomed.2024.109310","DOIUrl":"10.1016/j.compbiomed.2024.109310","url":null,"abstract":"<div><div>Acute type A aortic dissection remains a deadly and elusive condition, with risk factors such as hypertension, bicuspid aortic valves, and genetic predispositions. As existing guidelines for surgical intervention based solely on aneurysm diameter face scrutiny, there is a growing need to consider other predictors and parameters, including wall stress, in assessing dissection risk. Through our research, we aim to elucidate the biomechanical underpinnings of aortic dissection and provide valuable insights into its prediction and prevention.</div><div>We applied finite element analysis (FEA) to assess stress distribution on a rare dataset comprising computed tomography (CT) images obtained from eight patients at three stages of aortic dissection: pre-dissection (preD), post-dissection (postD), and post-repair (postR). Our findings reveal significant increases in both mean and peak aortic wall stresses during the transition from the preD state to the postD state, reflecting the mechanical impact of dissection. Surgical repair effectively restores aortic wall diameter to pre-dissection levels, documenting its effectiveness in mitigating further complications. Furthermore, we identified stress concentration regions within the aortic wall that closely correlated with observed dissection borders, offering insights into high-risk areas.</div><div>This study demonstrates the importance of considering biomechanical factors when assessing aortic dissection risk. Despite some limitations, such as uniform wall thickness assumptions and the absence of dynamic blood flow considerations, our patient-specific FEA approach provides valuable mechanistic insights into aortic dissection. These findings hold promise for improving predictive models and informing clinical decisions to enhance patient care.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109310"},"PeriodicalIF":7.0,"publicationDate":"2024-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142603392","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unraveling the relapse-associated landscape and individualized therapy in stage I lung adenocarcinoma based on immune and mitochondrial metabolism hallmarks via multi-omics analyses","authors":"Tao Zhan ,&nbsp;Luyao Wang ,&nbsp;Zewei Li ,&nbsp;Huijing Deng ,&nbsp;Liu Huang","doi":"10.1016/j.compbiomed.2024.109345","DOIUrl":"10.1016/j.compbiomed.2024.109345","url":null,"abstract":"<div><div>Lung adenocarcinoma (LUAD) is characterized by significant molecular heterogeneity and high recurrence rate even among stage I patients. There is an urgent quest for reliable biomarkers to recognize early-stage patients at high risk and guide potential treatment. Considering the pivotal role of immune and mitochondrial metabolic hallmarks in tumor initiation and progression, we rigorously included four independent cohorts of stage I LUAD patients with or without relapse. A consensus immune and mitochondrial metabolism genes-related signature (IMMS) is then constructed via 101 machine-learning combinations. IMMS classified all patients into high- and low-risk groups and exhibited a leading predicting accuracy compared with 70 previously published signatures. Subsequently, comprehensive analysis of the multi-omics data discovered elevated genomic heterogeneity, cancer stemness, metabolic reprogramming, immune escape, and tolerance to immune therapy in the high-risk group, which promotes the survival and proliferation of tumor cells. After the analysis of multiple drug databases, mitoxantrone is considered a candidate drug for stage I high-risk LUAD patients. The research of single-cell data further supported the tight association between IMMS and tumor cell characteristics. Overall, our study developed a novel signature and emphasized the role of immune escape and metabolic reprogramming hallmarks in recurrence, offering valuable insights into clinical prognosis, molecular mechanism, and individualized therapy for stage I LUAD patients.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109345"},"PeriodicalIF":7.0,"publicationDate":"2024-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142603406","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Synergistic biophysics and machine learning modeling to rapidly predict cardiac growth probability","authors":"Clara E. Jones,&nbsp;Pim J.A. Oomen","doi":"10.1016/j.compbiomed.2024.109323","DOIUrl":"10.1016/j.compbiomed.2024.109323","url":null,"abstract":"<div><div>Computational models that can predict growth and remodeling of the heart could have important clinical applications. However, the time it takes to calibrate and run current models while considering data uncertainty and variability makes them impractical for routine clinical use. This study aims to address this need by creating a computational framework to efficiently predict cardiac growth probability. We utilized a biophysics model to rapidly simulate cardiac growth following mitral valve regurgitation (MVR). Here we developed a two-tiered Bayesian History Matching approach augmented with Gaussian process emulators for efficient calibration of model parameters to align with growth outcomes within a 95% confidence interval. We first generated a synthetic data set to assess the accuracy of our framework, and the effect of changes in data uncertainty on growth predictions. We then calibrated our model to match baseline and chronic canine MVR data and used an independent data set to successfully validate the ability of our calibrated model to accurately predict cardiac growth probability. The combined biophysics and machine learning modeling framework we proposed in this study can be easily translated to predict patient-specific cardiac growth.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"184 ","pages":"Article 109323"},"PeriodicalIF":7.0,"publicationDate":"2024-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142603402","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"SARS-CoV-2 disrupts host gene networks: Unveiling key hub genes as potential therapeutic targets for COVID-19 management","authors":"Marta Majewska ,&nbsp;Mateusz Maździarz ,&nbsp;Katarzyna Krawczyk ,&nbsp;Łukasz Paukszto ,&nbsp;Karol G. Makowczenko ,&nbsp;Ewa Lepiarczyk ,&nbsp;Aleksandra Lipka ,&nbsp;Marta Wiszpolska ,&nbsp;Anna Górska ,&nbsp;Beata Moczulska ,&nbsp;Piotr Kocbach ,&nbsp;Jakub Sawicki ,&nbsp;Leszek Gromadziński","doi":"10.1016/j.compbiomed.2024.109343","DOIUrl":"10.1016/j.compbiomed.2024.109343","url":null,"abstract":"<div><h3>Purpose</h3><div>Although the end of COVID-19 as a public health emergency was declared on May 2023, still new cases of the infection are reported and the risk remains of new variants emerging that may cause new surges in cases and deaths. While clinical symptoms have been rapidly defined worldwide, the basic body responses and pathogenetic mechanisms acting in patients with SARS-CoV-2 infection over time until recovery or death require further investigation. The understanding of the molecular mechanisms underlying the development and course of the disease is essential in designing effective preventive and therapeutic approaches, and ultimately reducing mortality and disease spreading.</div></div><div><h3>Methods</h3><div>The current investigation aimed to identify the key genes engaged in SARS-CoV-2 infection. To achieve this goal high-throughput RNA sequencing of peripheral blood samples collected from healthy donors and COVID-19 patients was performed. The resulting sequence data were processed using a wide range of bioinformatics tools to obtain detailed modifications within five transcriptomic phenomena: expression of genes and long non-coding RNAs, alternative splicing, allel-specific expression and circRNA production. The <em>in silico</em> procedure was completed with a functional analysis of the identified alterations.</div></div><div><h3>Results</h3><div>The transcriptomic analysis revealed that SARS-CoV-2 has a significant impact on multiple genes encoding ribosomal proteins (RPs). Results show that these genes differ not only in terms of expression but also manifest biases in alternative splicing and ASE ratios. The integrated functional analysis exposed that RPs mostly affected pathways and processes related to infection—COVID-19 and NOD-like receptor signaling pathway, SARS-CoV-2-host interactions and response to the virus. Furthermore, our results linked the multiple intronic ASE variants and exonic circular RNA differentiations with SARS-CoV-2 infection, suggesting that these molecular events play a crucial role in mRNA maturation and transcription during COVID-19 disease.</div></div><div><h3>Conclusions</h3><div>By elucidating the genetic mechanisms induced by the virus, the current research provides significant information that can be employed to create new targeted therapeutic strategies for future research and treatment related to COVID-19. Moreover, the findings highlight potentially promising therapeutic biomarkers for early risk assessment of critically ill patients.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"183 ","pages":"Article 109343"},"PeriodicalIF":7.0,"publicationDate":"2024-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142581192","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}