Investigating the overlap of machine learning algorithms in the final results of RNA-seq analysis on gene expression estimation.

Advances in computer science in combination with the next-generation sequencing have introduced a new era in biology, enabling advanced state-of-the-art analysis of complex biological data. Bioinformatics is evolving as a union field between computer Science and biology, enabling the representation, storage, management, analysis and exploration of many types of data with a plethora of machine learning algorithms and computing tools. In this study, we used machine learning algorithms to detect differentially expressed genes between different types of cancer and showing the existence overlap to final results from RNA-sequencing analysis. The datasets were obtained from the National Center for Biotechnology Information resource. Specifically, dataset GSE68086 which corresponds to PMID:200,068,086. This dataset consists of 171 blood platelet samples collected from patients with six different tumors and healthy individuals. All steps for RNA-sequencing analysis (preprocessing, read alignment, transcriptome reconstruction, expression quantification and differential expression analysis) were followed. Machine Learning- based Random Forest and Gradient Boosting algorithms were applied to predict significant genes. The Rstudio statistical tool was used for the analysis.

TarBase-v9.0 extends experimentally supported miRNA-gene interactions to cell-types and virally encoded miRNAs.

TarBase is a reference database dedicated to produce, curate and deliver high quality experimentally-supported microRNA (miRNA) targets on protein-coding transcripts. In its latest version (v9.0, https://dianalab.e-ce.uth.gr/tarbasev9), it pushes the envelope by introducing virally-encoded miRNAs, interactions leading to target-directed miRNA degradation (TDMD) events and the largest collection of miRNA-gene interactions to date in a plethora of experimental settings, tissues and cell-types. It catalogues ∼6 million entries, comprising ∼2 million unique miRNA-gene pairs, supported by 37 experimental (high- and low-yield) protocols in 172 tissues and cell-types. Interactions are annotated with rich metadata including information on genes/transcripts, miRNAs, samples, experimental contexts and publications, while millions of miRNA-binding locations are also provided at cell-type resolution. A completely re-designed interface with state-of-the-art web technologies, incorporates more features, and allows flexible and ingenious use. The new interface provides the capability to design sophisticated queries with numerous filtering criteria including cell lines, experimental conditions, cell types, experimental methods, species and/or tissues of interest. Additionally, a plethora of fine-tuning capacities have been integrated to the platform, offering the refinement of the returned interactions based on miRNA confidence and expression levels, while boundless local retrieval of the offered interactions and metadata is enabled.

Analysing sentiment change detection of Covid-19 tweets.

The Covid-19 pandemic made a significant impact on society, including the widespread implementation of lockdowns to prevent the spread of the virus. This measure led to a decrease in face-to-face social interactions and, as an equivalent, an increase in the use of social media platforms, such as Twitter. As part of Industry 4.0, sentiment analysis can be exploited to study public attitudes toward future pandemics and sociopolitical situations in general. This work presents an analysis framework by applying a combination of natural language processing techniques and machine learning algorithms to classify the sentiment of each tweet as positive, or negative. Through extensive experimentation, we expose the ideal model for this task and, subsequently, utilize sentiment predictions to perform time series analysis over the course of the pandemic. In addition, a change point detection algorithm was applied in order to identify the turning points in public attitudes toward the pandemic, which were validated by cross-referencing the news report at that particular period of time. Finally, we study the relationship between sentiment trends on social media and, news coverage of the pandemic, providing insights into the public's perception of the pandemic and its influence on the news.

Greek Hospital Data Mining and Analysis.

Monitoring the performance of hospitals is a crucial issue related both with the quality of healthcare services and with country's economy. An easy and trustful way of evaluating health systems is through key performance indicators (KPIs). Such indicators are widely used for the identification of gaps in the quality or efficiency of the services provided. The main aim of this study is the analysis of the financial and operational indicators at hospitals in the 3rd and 5th Healthcare Regions of Greece. In addition, through cluster analysis and data visualization we attempt to uncover hidden patterns that may lie within our data. The results of the study support the need for re-evaluation of the assessment methodology of Greek hospitals to identify the weaknesses in the system, while evidently unsupervised learning exposes the potential of group-based decision making.

The development and validation of the pandemic medication-assisted treatment questionnaire for the assessment of pandemic crises impact on medication management and administration for patients with opioid use disorders.

Pandemic and the globally applied restriction measures mainly affect vulnerable population groups, such as patients with opioid use disorders. Towards inhibiting SARS-Cov-2 spread, the medication-assisted treatment (MAT) programs follow strategies targeting the reduction of in-person psychosocial interventions and an increase of take-home doses. However, there is no available instrument to examine the impact of such modifications on diverse health aspects of patients under MAT. The aim of this study was to develop and validate the PANdemic Medication-Assisted Treatment Questionnaire (PANMAT/Q) to address the pandemic effect on the management and administration of MAT. In total, 463 patients under ΜΑΤ participated. Our findings indicate that PANMAT/Q has been successfully validated exerting reliability and validity. It can be completed within approximately 5 min, and its implementation in research settings is advocated. PANMAT/Q could serve as a useful tool to identify the needs of patients under MAT being at high risk of relapse and overdose.

A divisive hierarchical clustering methodology for enhancing the ensemble prediction power in large scale population studies: the ATHLOS project.

The ATHLOS cohort is composed of several harmonized datasets of international groups related to health and aging. As a result, the Healthy Aging index has been constructed based on a selection of variables from 16 individual studies. In this paper, we consider additional variables found in ATHLOS and investigate their utilization for predicting the Healthy Aging index. For this purpose, motivated by the volume and diversity of the dataset, we focus our attention upon data clustering, where unsupervised learning is utilized to enhance prediction power. Thus we show the predictive utility of exploiting hidden data structures. In addition, we demonstrate that imposed computation bottlenecks can be surpassed when using appropriate hierarchical clustering, within a clustering for ensemble classification scheme, while retaining prediction benefits. We propose a complete methodology that is evaluated against baseline methods and the original concept. The results are very encouraging suggesting further developments in this direction along with applications in tasks with similar characteristics. A straightforward open source implementation for the R project is also provided (https://github.com/Petros-Barmpas/HCEP). Supplementary Information: The online version contains supplementary material available at 10.1007/s13755-022-00171-1.

Atlantes: Automated Health Related & COVID-19 Data Management for Use in Predictive Models.

The scientific community, having turned its interest, almost entirely, to the treatment and understanding of COVID-19, is constantly striving to collect and use data from the countless available sources. That data, however, is scattered, not designed to be combined, collected in different time periods and their volume is constantly increasing. In this paper, we present an automated methodology that collects, refines, groups and combines data for a large number of countries. Most of these data resources are directly related to COVID-19 but we also choose to include other types of variables for each country, which may be of particular interest for researchers working in understanding the COVID-19 pandemic. The presented methodology unifies critical information regarding the pandemic. It is implemented in Python, provided as a simple script that extracts data, in the form of a daily time series, in a short period of time, directly available to be incorporated for analysis.

Knee Injury Detection Using Deep Learning on MRI Studies: A Systematic Review.

The improved treatment of knee injuries critically relies on having an accurate and cost-effective detection. In recent years, deep-learning-based approaches have monopolized knee injury detection in MRI studies. The aim of this paper is to present the findings of a systematic literature review of knee (anterior cruciate ligament, meniscus, and cartilage) injury detection papers using deep learning. The systematic review was carried out following the PRISMA guidelines on several databases, including PubMed, Cochrane Library, EMBASE, and Google Scholar. Appropriate metrics were chosen to interpret the results. The prediction accuracy of the deep-learning models for the identification of knee injuries ranged from 72.5-100%. Deep learning has the potential to act at par with human-level performance in decision-making tasks related to the MRI-based diagnosis of knee injuries. The limitations of the present deep-learning approaches include data imbalance, model generalizability across different centers, verification bias, lack of related classification studies with more than two classes, and ground-truth subjectivity. There are several possible avenues of further exploration of deep learning for improving MRI-based knee injury diagnosis. Explainability and lightweightness of the deployed deep-learning systems are expected to become crucial enablers for their widespread use in clinical practice.

An Online Information Tool for Diabetic Retinopathy.

Regardless of the type of diabetes, patients with diabetes are 25 times more likely to develop vision problems or even blindness than non-diabetics. Diabetic Retinopathy is the most common cause of new cases of blindness in adults. The aim of this paper is to present a pilot online tool to provide information regarding the Diabetic Retinopathy. The tool was developed using a Content Management System. To compile the content of the website, a literature review was conducted. The online information tool is addressed to all potential stakeholders on this subject, for the provision of knowledge and targeted information according to their information needs. The online tool also aims to raise the public awareness about the Diabetic Retinopathy and health promotion.

Systems Toxicology Approach to Identifying Paracetamol Overdose.

Paracetamol (acetaminophen (APAP)) is one of the most commonly used analgesics in the United Kingdom and the United States. However, exceeding the maximum recommended dose can cause serious liver injury and even death. Promising APAP toxicity biomarkers are thought to add value to those used currently and clarification of the functional relationships between these biomarkers and liver injury would aid clinical implementation of an improved APAP toxicity identification framework. The framework currently used to define an APAP overdose is highly dependent upon time since ingestion and initial dose; information that is often highly unpredictable. A pharmacokinetic/pharmacodynamic (PK/PD) APAP model has been built in order to understand the relationships between a panel of biomarkers and APAP dose. Visualization and statistical tools have been used to predict initial APAP dose and time since administration. Additionally, logistic regression analysis has been applied to histology data to provide a prediction of the probability of liver injury.