Lightweight neural architectures to improve COVID-19 identification

The COVID-19 pandemic has had a global impact, transforming how we manage infectious diseases and interact socially. Researchers from various fields have worked tirelessly to develop vaccines on an unprecedented scale, while different countries have developed various sanitary protocols to deal with more contagious variants. Machine learning-assisted diagnosis has emerged as a powerful tool that can help health professionals deliver faster and more accurate outcomes. However, medical systems that rely on deep learning often require extensive data, which may be impractical for real-world applications. This paper compares lightweight neural architectures for COVID-19 identification using chest X-rays, highlighting the strengths and weaknesses of each approach. Additionally, a web tool has been developed that accepts chest computer tomography images and outputs the probability of COVID-19 infection along with a heatmap of the regions used by the intelligent system to make this determination. The experiments indicate that most lightweight architectures considered in the study can identify COVID-19 correctly, but further investigation is necessary. Lightweight neural architectures show promise in computer-aided COVID-19 diagnosis using chest X-rays, but they did not reach accuracy rates above 88%, which is necessary for medical applications. These findings suggest that additional research is necessary to improve the accuracy of lightweight models and make them practical for real-world use. Copyright © 2023 Hassan, AlQahtani, Alelaiwi and Papa.

PolyX2: Fast Detection of Homorepeats in Large Protein Datasets.

Homorepeat sequences, consecutive runs of identical amino acids, are prevalent in eukaryotic proteins. It has become necessary to annotate and evaluate this feature in entire proteomes. The definition of what constitutes a homorepeat is not fixed, and different research approaches may require different definitions; therefore, flexible approaches to analyze homorepeats in complete proteomes are needed. Here, we present polyX2, a fast, simple but tunable script to scan protein datasets for all possible homorepeats. The user can modify the length of the window to scan, the minimum number of identical residues that must be found in the window, and the types of homorepeats to be found.

E-Volve: understanding the impact of mutations in SARS-CoV-2 variants spike protein on antibodies and ACE2 affinity through patterns of chemical interactions at protein interfaces.

Background: The SARS-CoV-2 pandemic reverberated, posing health and social hygiene obstacles throughout the globe. Mutant lineages of the virus have concerned scientists because of convergent amino acid alterations, mainly on the viral spike protein. Studies have shown that mutants have diminished activity of neutralizing antibodies and enhanced affinity with its human cell receptor, the ACE2 protein. Methods: Hence, for real-time measuring of the impacts caused by variant strains in such complexes, we implemented E-Volve, a tool designed to model a structure with a list of mutations requested by users and return analyses of the variant protein. As a proof of concept, we scrutinized the spike-antibody and spike-ACE2 complexes formed in the variants of concern, B.1.1.7 (Alpha), B.1.351 (Beta), and P.1 (Gamma), by using contact maps depicting the interactions made amid them, along with heat maps to quantify these major interactions. Results: The results found in this study depict the highly frequent interface changes made by the entire set of mutations, mainly conducted by N501Y and E484K. In the spike-Antibody complex, we have noticed alterations concerning electrostatic surface complementarity, breaching essential sites in the P17 and BD-368-2 antibodies. Alongside, the spike-ACE2 complex has presented new hydrophobic bonds. Discussion: Molecular dynamics simulations followed by Poisson-Boltzmann calculations corroborate the higher complementarity to the receptor and lower to the antibodies for the K417T/E484K/N501Y (Gamma) mutant compared to the wild-type strain, as pointed by E-Volve, as well as an intensification of this effect by changes at the protein conformational equilibrium in solution. A local disorder of the loop α1'/ß1', as well its possible effects on the affinity to the BD-368-2 antibody were also incorporated to the final conclusions after this analysis. Moreover, E-Volve can depict the main alterations in important biological structures, as shown in the SARS-CoV-2 complexes, marking a major step in the real-time tracking of the virus mutant lineages. E-Volve is available at http://bioinfo.dcc.ufmg.br/evolve.

A Clinical Decision Web to Predict ICU Admission or Death for Patients Hospitalised with COVID-19 Using Machine Learning Algorithms.

The purpose of the study was to build a predictive model for estimating the risk of ICU admission or mortality among patients hospitalized with COVID-19 and provide a user-friendly tool to assist clinicians in the decision-making process. The study cohort comprised 3623 patients with confirmed COVID-19 who were hospitalized in the SALUD hospital network of Aragon (Spain), which includes 23 hospitals, between February 2020 and January 2021, a period that includes several pandemic waves. Up to 165 variables were analysed, including demographics, comorbidity, chronic drugs, vital signs, and laboratory data. To build the predictive models, different techniques and machine learning (ML) algorithms were explored: multilayer perceptron, random forest, and extreme gradient boosting (XGBoost). A reduction dimensionality procedure was used to minimize the features to 20, ensuring feasible use of the tool in practice. Our model was validated both internally and externally. We also assessed its calibration and provide an analysis of the optimal cut-off points depending on the metric to be optimized. The best performing algorithm was XGBoost. The final model achieved good discrimination for the external validation set (AUC = 0.821, 95% CI 0.787-0.854) and accurate calibration (slope = 1, intercept = -0.12). A cut-off of 0.4 provides a sensitivity and specificity of 0.71 and 0.78, respectively. In conclusion, we built a risk prediction model from a large amount of data from several pandemic waves, which had good calibration and discrimination ability. We also created a user-friendly web application that can aid rapid decision-making in clinical practice.

An Evolutionary Portrait of the Progenitor SARS-CoV-2 and Its Dominant Offshoots in COVID-19 Pandemic.

Global sequencing of genomes of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has continued to reveal new genetic variants that are the key to unraveling its early evolutionary history and tracking its global spread over time. Here we present the heretofore cryptic mutational history and spatiotemporal dynamics of SARS-CoV-2 from an analysis of thousands of high-quality genomes. We report the likely most recent common ancestor of SARS-CoV-2, reconstructed through a novel application and advancement of computational methods initially developed to infer the mutational history of tumor cells in a patient. This progenitor genome differs from genomes of the first coronaviruses sampled in China by three variants, implying that none of the earliest patients represent the index case or gave rise to all the human infections. However, multiple coronavirus infections in China and the United States harbored the progenitor genetic fingerprint in January 2020 and later, suggesting that the progenitor was spreading worldwide months before and after the first reported cases of COVID-19 in China. Mutations of the progenitor and its offshoots have produced many dominant coronavirus strains that have spread episodically over time. Fingerprinting based on common mutations reveals that the same coronavirus lineage has dominated North America for most of the pandemic in 2020. There have been multiple replacements of predominant coronavirus strains in Europe and Asia as well as continued presence of multiple high-frequency strains in Asia and North America. We have developed a continually updating dashboard of global evolution and spatiotemporal trends of SARS-CoV-2 spread (http://sars2evo.datamonkey.org/).