Search

Background: The coronavirus disease-2019 (COVID-19) pandemic, caused by severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2), has urgently necessitated effective therapeutic solutions, with a focus on rapidly identifying and classifying potential small-molecule drugs. Given traditional methods’ labor-intensive and time-consuming nature, deep learning has emerged as an essential tool for efficiently processing and extracting insights from complex biological data. Aims: To utilize deep learning techniques, particularly deep neural networks (DNN) enhanced with the synthetic minority oversampling technique (SMOTE), to enhance the classification of binding activities in anti-SARS-CoV-2 molecules across various bioassays. Methods: We used 11 bioassay datasets covering various SARS-CoV-2 interactions and inhibitory mechanisms. These assays ranged from spike-ACE2 protein-protein interaction to ACE2 enzymatic activity and 3CL enzymatic activity. To address the prevalent class imbalance in these datasets, the SMOTE technique was employed to generate new samples for the minority class. In our model-building approach, we divided the dataset into 80% training and 20% test sets, reserving 10% of the training set for validation. Our approach involved employing a DNN that integrates ReLU and sigmoid activation functions, incorporates batch normalization, and uses Adam optimization. The hyperparameters and architecture of the DNN were optimized through various tests on layers, minibatch sizes, epoch sizes, and learning rates. A 40% dropout rate was incorporated to mitigate overfitting. For model evaluation, we computed performance metrics, such as balanced accuracy (BACC), precision, recall, F1 score, Matthews’ correlation coefficient (MCC), and area under the curve (AUC). Results: The performance of the DNN across 11 bioassay test sets revealed varying outcomes, significantly influenced by the ratios of active-to-inactive compounds. Assays, such as AlphaLISA and CoV-PPE, demonstrated robust performance across various metrics, including BACC, precision, recall, and AUC, when configured with more balanced ratios (1:3 and 1:1, respectively). This suggests the effective identification of active compounds in both cases. In contrast, assays with higher imbalance ratios, such as 3CL (1:38) and cytopathic effect (1:15), demonstrated higher recall but lower precision, highlighting challenges in accurately identifying active compounds among numerous inactive compounds. However, even in these challenging settings, the model achieved favorable BACC and recall scores. Overall, the DNN model generally performed well, as indicated by the BACC, MCC, and AUC values, especially when considering the degree of dataset imbalance in each assay. Conclusion: This study demonstrates the significant impact of deep learning, particularly DNN models enhanced with SMOTE, in improving the identification of active compounds in bioassay datasets for COVID-19 drug discovery, outperforming traditional machine learning models. Furthermore, this study highlights the efficacy of advanced computational techniques in addressing high-throughput screening data imbalances.

This study aimed to compare the efficacy of percutaneous microwave ablation therapy (MWAT) and laparoscopic partial nephrectomy (LPN) in early-stage renal cell carcinoma (RCC) classified as T1a; a retrospective analysis was conducted on patients treated between January 2017 and November 2023. Oncological outcomes, radiological recurrence, length of stay (LOS), and costs were evaluated. The study included 110 patients, with no significant differences between the two groups regarding residual tumors, local tumor progression, and disease-free survival rates (p > 0.05). The LPN group showed significantly lower pre/postoperative serum urea and creatinine and higher estimated glomerular filtration rate values, whereas the MWA group experienced significantly lower mean costs, complication rates, LOS in the hospital, and procedure durations (p ≤ 0.05). However, post-procedure residual tumors and local tumor progression rates did not differ significantly between the LPN and MWAT groups (p > 0.05). MWAT is as effective as LPN for T1a RCC lesions. In addition, MWAT has lower costs than LPN and is a cost-effective treatment method. Therefore, MWAT minimizes hospital stay and complications and since the oncological results are similar to LPN, it might be considered as the first choice of treatment in young patients.

Background: Primary Sjögren’s syndrome is a chronic inflammatory autoimmune disease. Minor salivary gland biopsy is the gold standard for the diagnosis of primary Sjögren’s syndrome. Superb microvascular imaging, power Doppler ultrasound, and color Doppler of the salivary glands represent non-invasive, non-irradiating modality for evaluating the vascularity of the salivary glands in the diagnosis and follow-up of primary Sjögren’s syndrome. Aims: To evaluate the efficacy of superb microvascular imaging and vascularity index in salivary glands for the sonographic diagnosis of primary Sjögren’s syndrome. Study Design: Prospective case-control study. Methods: Twenty participants with primary Sjögren’s syndrome and 20 healthy subjects were included in the study. Both parotid glands and submandibular glands were evaluated by superb microvascular imaging, power Doppler ultrasound, and color Doppler. The diagnostic accuracy of superb microvascular imaging was compared using these techniques. Results: In the patient group, the vascularity index values of superb microvascular imaging in parotid glands and submandibular glands were 3.5±1.66, 5.06±1.94, respectively. While the same values were 1.0±0.98 and 2.44±1.34 in the control group (p≤0.001). In the patient group, the vascularity index values of power Doppler ultrasound in parotid glands and submandibular glands were 1.3±1.20 and 2.59±1.82, respectively. While the same values were 0.3±0.32 and 0.85±0.68 in the control group (p≤0.001). The superb microvascular imaging vascularity index cut-off value for the diagnosis of primary Sjögren’s syndrome in parotid glands that maximizes the accuracy was 1.85 (area under the curve: 0.906; 95% confidence interval: 0.844, 0.968), and its sensitivity and specificity were 87.5% and 72.5%, respectively. While the superb microvascular imaging vascularity index cut-off value for the diagnosis of primary Sjögren’s syndrome in submandibular gland that maximizes the accuracy was 3.35 (area under the curve: 0.873; 95% confidence interval: 0.800, 0.946), its sensitivity and specificity were 82.5% and 70%, respectively. Conclusion: Superb microvascular imaging with high reproducibility of the vascularity index has a higher sensitivity and specificity than the power Doppler ultrasound in the diagnosis of primary Sjögren’s syndrome. It can be a noninvasive technique in the diagnosis of primary Sjögren’s syndrome when used with clinical, laboratory and other imaging methods.

Background and objectives: Cardiac magnetic resonance (CMR) imaging is the gold standard method for the detection of ventricular volumes and myocardial edema/scar. Transthoracic echocardiography (TTE) imaging is primarily used in the evaluation of cardiac functions and chamber dimensions. This study aims to investigate whether the chamber diameter measurements are concordant with each other in the same patient group who underwent TTE and CMR. Materials and Methods: The study included 41 patients who underwent TTE and CMR imaging. Ventricular and atrial diameter measurements from TTE-derived standard parasternal long axis and apical four-chamber views and CMR-derived three- and four-chamber views were recorded. The concordance between the two methods was compared using intra-class correlation coefficients (ICC) and Bland−Altman plots. Results: Of the patients, 25 (61%) were male and the mean age was 48.12 ± 16.79. The mean ICC for LVDD between CMR observers was 0.957 (95% CI: 0.918−0.978), while the mean ICC between CMR and TTE measurements were 0.849 (95% CI: 0.709−0.922) and 0.836 (95% CI: 0.684−0.915), respectively. The mean ICC for the right ventricle between CMR observers was 0.985 (95% CI: 0.971−0.992), while the mean ICC between CMR and TTE measurements were 0.869 (95% CI: 0.755−0.930) and 0.892 (95% CI: 0.799−0.942), respectively. Passing−Bablok Regression and Bland−Altman plots indicated high concordance between the two methods. Conclusions: TTE and CMR indicated high concordance in chamber diameter measurements for which the CMR should be considered in patients for whom optimal evaluation with TTE could not be performed due to their limitations.

Background and objectives: Despite being within the normal reference range, changes in thyroid stimulating hormone (TSH) levels have negative effects on the cardiovascular system. The majority of patients admitted to hospital with acute coronary syndrome (ACS) are euthyroid. The aim of this study was to investigate the effect of TSH level on the prognosis of in-hospital and follow-up periods of euthyroid ACS patients. Materials and Methods: A total of 629 patients with acute coronary syndrome without thyroid dysfunction were included in the study. TSH levels of patients were 0.3–5.33 uIU/mL. Patients were divided into three TSH tertiles: TSH level between (1) 0.3 uIU/mL and n = 209), (2) 0.90 uIU/mL and n = 210), and (3) 1.60 uIU/mL and 5.33 uIU/mL (n = 210). Demographic, clinical laboratory, and angiographic characteristics were compared between groups in terms of in-hospital and follow-up prognosis. Results: Mean age was 63.42 ± 12.5, and 73.9% were male. There was significant difference between tertiles in terms of TSH level at admission (p < 0.001), the severity of coronary artery disease (p = 0.024), in-hospital mortality (p < 0.001), in-hospital major hemorrhage (p = 0.005), total adverse clinical event (p = 0.03), follow-up mortality (p = 0.022), and total mortality (p < 0.001). In multivariate logistic regression analysis, the high–normal TSH tertile was found to be cumulative mortality increasing factor (OR = 6.307, 95%; CI: 1.769–22.480; p = 0.005) during the 6-month follow-up period after hospitalization and discharge. Conclusions: High–normal TSH tertile during hospital admission in euthyroid ACS patients is an independent predictor of total mortality during the 6-month follow-up period after hospitalization and discharge.

Background and objectives: Cranial magnetic resonance imaging findings of patients considered to be cryptogenic stroke may be useful in determining the clinical and prognostic significance of arrhythmias, such as atrial premature beats and atrial run attacks, that are frequently encountered in rhythm Holter analysis. This study was conducted to investigate the relationship between short atrial runs and frequent premature atrial contractions detected in Holter monitors and infarct distributions in cranial magnetic resonance imaging of patients diagnosed with cryptogenic stroke. Materials and Methods: We enrolled the patients with acute ischemic stroke whose etiology were undetermined. We divided the patients in two groups according to diffusion-weighted magnetic resonance imaging as single or multiple vascular territory acute infarcts. The demographic, clinical, laboratory, echocardiographic, and rhythm Holter analyses were compared. Results: The study investigated 106 patients diagnosed with cryptogenic stroke. Acute cerebral infarctions were detected in 31% of the investigated patients in multiple territories and in 69% in a single territory. In multivariate logistic regression analysis, the total premature atrial contraction count (OR = 1.002, 95% CI: 1.001⁻1.004, p = 0.001) and short atrial run count (OR = 1.086, 95% CI: 1.021⁻1.155, p = 0.008) were found as independent variables that could distinguish between infarctions in a single or in multiple vascular territories. Conclusions: Rhythm Holter monitoring of patients with infarcts detected in multiple vascular territories showed significantly higher premature atrial contractions and short atrial run attacks. More effort should be devoted to the identification of cardioembolic etiology in cryptogenic stroke patients with concurrent acute infarcts in the multiple vascular territories of the brain.

Background: Published reports of randomized controlled trials (RCTs) are not compliant with the CONSORT checklist as much as they should.Objective: To assess the quality, in terms of the level to which they are compliant with the CONSORT checklist, of abstracts of RCTs published in general medical journals in the Balkan region.Methods: Two observers assessed the abstracts of RCTs published in five general medical journals of the Balkan region between 2012 and 2018 to determine the level to which the abstracts were compliant with the 16-item CONSORT abstracts checklist.Results: Of the 183 studies that were identified for evaluation, 124 (67.8%) were excluded from the evaluation. The average compliance level was 44.5% (95% CI: 41.9%–47.1%), the lowest being that for randomization (1.7%), funding (1.7%),numbers analysed (11.0%), blinding (13.6%), and trial registration (18.6%). However, the compliance level was very high for conclusions (99.2%), objectives (96.6%), interventions (95.8%), and primary outcomes (83.9%). The length of the abstract (word count) and the level of compliance were positively correlated (rs = 0.43; p = 0.001). Abstracts of trials published in journals that endorse CONSORT in their publication policies were more compliant than those published in other journals (47.5 ± 10.4 versus 40.8 ± 8.0, p = 0.024).Conclusion: The overall level of compliance with the CONSORT checklist was below 50%. To improve the quality of abstracts of RCTs, authors should be encouraged to use the CONSORT checklist, and editors should check compliance with the CONSORT guidelines as part the publishing workflow.

The Protein Data Bank (PDB) is the single worldwide archive of experimentally-determined three-dimensional (3D) structures of proteins and nucleic acids. As of January 2017, the PDB housed more than 125,000 structures and was growing by more than 11,000 structures annually. Since the 3D structure of a protein is vital to understand the mechanisms of biological processes, diseases, and drug design, correct oligomeric assembly information is of critical importance. Unfortunately, the biologically relevant oligomeric form of a 3D structure is not directly obtainable by X-ray crystallography, whilst in solution methods (NMR or single particle EM) it is known from the experiment. Instead, this information may be provided by the PDB Depositor as metadata coming from additional experiments, be inferred by sequence-sequence comparisons with similar proteins of known oligomeric state, or predicted using software, such as PISA (Proteins, Interfaces, Structures and Assemblies) or EPPIC (Evolutionary Protein Protein Interface Classifier). Despite significant efforts by professional PDB Biocurators during data deposition, there remain a number of structures in the archive with incorrect quaternary structure descriptions (or annotations). Further investigation is, therefore, needed to evaluate the correctness of quaternary structure annotations. In this study, we aim to identify the most probable oligomeric states for proteins represented in the PDB. Our approach evaluated the performance of four independent prediction methods, including text mining of primary publications, inference from homologous protein structures, and two computational methods (PISA and EPPIC). Aggregating predictions to give consensus results outperformed all four of the independent prediction methods, yielding 83% correct, 9% wrong, and 8% inconclusive predictions, when tested with a well-curated benchmark dataset. We have developed a freely-available web-based tool to make this approach accessible to researchers and PDB Biocurators (http://quatstruct.rcsb.org/).

RNA-Seq is a recent and efficient technique that uses the capabilities of next-generation sequencing technology for characterizing and quantifying transcriptomes. One important task using gene-expression data is to identify a small subset of genes that can be used to build diagnostic classifiers particularly for cancer diseases. Microarray based classifiers are not directly applicable to RNA-Seq data due to its discrete nature. Overdispersion is another problem that requires careful modeling of mean and variance relationship of the RNA-Seq data. In this study, we present voomDDA classifiers: variance modeling at the observational level (voom) extensions of the nearest shrunken centroids (NSC) and the diagonal discriminant classifiers. VoomNSC is one of these classifiers and brings voom and NSC approaches together for the purpose of gene-expression based classification. For this purpose, we propose weighted statistics and put these weighted statistics into the NSC algorithm. The VoomNSC is a sparse classifier that models the mean-variance relationship using the voom method and incorporates voom’s precision weights into the NSC classifier via weighted statistics. A comprehensive simulation study was designed and four real datasets are used for performance assessment. The overall results indicate that voomNSC performs as the sparsest classifier. It also provides the most accurate results together with power-transformed Poisson linear discriminant analysis, rlog transformed support vector machines and random forests algorithms. In addition to prediction purposes, the voomNSC classifier can be used to identify the potential diagnostic biomarkers for a condition of interest. Through this work, statistical learning methods proposed for microarrays can be reused for RNA-Seq data. An interactive web application is freely available at http://www.biosoft.hacettepe.edu.tr/voomDDA/.

RNA sequencing (RNA-Seq) is a powerful technique for the gene-expression profiling of organisms that uses the capabilities of next-generation sequencing technologies. Developing gene-expression-based classification algorithms is an emerging powerful method for diagnosis, disease classification and monitoring at molecular level, as well as providing potential markers of diseases. Most of the statistical methods proposed for the classification of gene-expression data are either based on a continuous scale (eg. microarray data) or require a normal distribution assumption. Hence, these methods cannot be directly applied to RNA-Seq data since they violate both data structure and distributional assumptions. However, it is possible to apply these algorithms with appropriate modifications to RNA-Seq data. One way is to develop count-based classifiers, such as Poisson linear discriminant analysis and negative binomial linear discriminant analysis. Another way is to bring the data closer to microarrays and apply microarray-based classifiers. In this study, we compared several classifiers including PLDA with and without power transformation, NBLDA, single SVM, bagging SVM (bagSVM), classification and regression trees (CART), and random forests (RF). We also examined the effect of several parameters such as overdispersion, sample size, number of genes, number of classes, differential-expression rate, and the transformation method on model performances. A comprehensive simulation study is conducted and the results are compared with the results of two miRNA and two mRNA experimental datasets. The results revealed that increasing the sample size, differential-expression rate and decreasing the dispersion parameter and number of groups lead to an increase in classification accuracy. Similar with differential-expression studies, the classification of RNA-Seq data requires careful attention when handling data overdispersion. We conclude that, as a count-based classifier, the power transformed PLDA and, as a microarray-based classifier, vst or rlog transformed RF and SVM classifiers may be a good choice for classification. An R/BIOCONDUCTOR package, MLSeq, is freely available at https://www.bioconductor.org/packages/release/bioc/html/MLSeq.html.

Virtual screening is an important step in early-phase of drug discovery process. Since there are thousands of compounds, this step should be both fast and effective in order to distinguish drug-like and nondrug-like molecules. Statistical machine learning methods are widely used in drug discovery studies for classification purpose. Here, we aim to develop a new tool, which can classify molecules as drug-like and nondrug-like based on various machine learning methods, including discriminant, tree-based, kernel-based, ensemble and other algorithms. To construct this tool, first, performances of twenty-three different machine learning algorithms are compared by ten different measures, then, ten best performing algorithms have been selected based on principal component and hierarchical cluster analysis results. Besides classification, this application has also ability to create heat map and dendrogram for visual inspection of the molecules through hierarchical cluster analysis. Moreover, users can connect the PubChem database to download molecular information and to create two-dimensional structures of compounds. This application is freely available through www.biosoft.hacettepe.edu.tr/MLViS/.