Determination of prognostic markers for COVID-19 disease severity using routine blood tests and machine learning.

The need for the identification of risk factors associated to COVID-19 disease severity remains urgent. Patients' care and resource allocation can be potentially different and are defined based on the current classification of disease severity. This classification is based on the analysis of clinical parameters and routine blood tests, which are not standardized across the globe. Some laboratory test alterations have been associated to COVID-19 severity, although these data are conflicting partly due to the different methodologies used across different studies. This study aimed to construct and validate a disease severity prediction model using machine learning (ML). Seventy-two patients admitted to a Brazilian hospital and diagnosed with COVID-19 through RT-PCR and/or ELISA, and with varying degrees of disease severity, were included in the study. Their electronic medical records and the results from daily blood tests were used to develop a ML model to predict disease severity. Using the above data set, a combination of five laboratorial biomarkers was identified as accurate predictors of COVID-19 severe disease with a ROC-AUC of 0.80 ​±â€‹ 0.13. Those biomarkers included prothrombin activity, ferritin, serum iron, ATTP and monocytes. The application of the devised ML model may help rationalize clinical decision and care.

An artificial neural network model to predict structure-based protein-protein free energy of binding from Rosetta-calculated properties.

The prediction of the free energy (ΔG) of binding for protein-protein complexes is of general scientific interest as it has a variety of applications in the fields of molecular and chemical biology, materials science, and biotechnology. Despite its centrality in understanding protein association phenomena and protein engineering, the ΔG of binding is a daunting quantity to obtain theoretically. In this work, we devise a novel Artificial Neural Network (ANN) model to predict the ΔG of binding for a given three-dimensional structure of a protein-protein complex with Rosetta-calculated properties. Our model was tested using two data sets, and it presented a root-mean-square error ranging from 1.67 kcal mol-1 to 2.45 kcal mol-1, showing a better performance compared to the available state-of-the-art tools. Validation of the model for a variety of protein-protein complexes is showcased.

Inhibition of 3-Hydroxykynurenine Transaminase from Aedes aegypti and Anopheles gambiae: A Mosquito-Specific Target to Combat the Transmission of Arboviruses.

Arboviral infections such as Zika, chikungunya, dengue, and yellow fever pose significant health problems globally. The population at risk is expanding with the geographical distribution of the main transmission vector of these viruses, the Aedes aegypti mosquito. The global spreading of this mosquito is driven by human migration, urbanization, climate change, and the ecological plasticity of the species. Currently, there are no specific treatments for Aedes-borne infections. One strategy to combat different mosquito-borne arboviruses is to design molecules that can specifically inhibit a critical host protein. We obtained the crystal structure of 3-hydroxykynurenine transaminase (AeHKT) from A. aegypti, an essential detoxification enzyme of the tryptophan metabolism pathway. Since AeHKT is found exclusively in mosquitoes, it provides the ideal molecular target for the development of inhibitors. Therefore, we determined and compared the free binding energy of the inhibitors 4-(2-aminophenyl)-4-oxobutyric acid (4OB) and sodium 4-(3-phenyl-1,2,4-oxadiazol-5-yl)butanoate (OXA) to AeHKT and AgHKT from Anopheles gambiae, the only crystal structure of this enzyme previously known. The cocrystallized inhibitor 4OB binds to AgHKT with K i of 300 µM. We showed that OXA binds to both AeHKT and AgHKT enzymes with binding energies 2-fold more favorable than the crystallographic inhibitor 4OB and displayed a 2-fold greater residence time τ upon binding to AeHKT than 4OB. These findings indicate that the 1,2,4-oxadiazole derivatives are inhibitors of the HKT enzyme not only from A. aegypti but also from A. gambiae.

Immune evasion of SARS-CoV-2 variants of concern is driven by low affinity to neutralizing antibodies.

SARS-CoV-2 VOC immune evasion is mainly due to lower cross-reactivity from previously elicited class I/II neutralizing antibodies, while increased affinity to hACE2 plays a minor role. The affinity between antibodies and VOCs is impacted by remodeling of the electrostatic surface potential of the Spike RBDs. The P.3 variant is a putative VOC.

The ongoing evolution of variants of concern and interest of SARS-CoV-2 in Brazil revealed by convergent indels in the amino (N)-terminal domain of the spike protein.

Mutations at both the receptor-binding domain (RBD) and the amino (N)-terminal domain (NTD) of the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Spike (S) glycoprotein can alter its antigenicity and promote immune escape. We identified that SARS-CoV-2 lineages circulating in Brazil with mutations of concern in the RBD independently acquired convergent deletions and insertions in the NTD of the S protein, which altered the NTD antigenic-supersite and other predicted epitopes at this region. Importantly, we detected the community transmission of different P.1 lineages bearing NTD indels ∆69-70 (which can impact several SARS-CoV-2 diagnostic protocols), ∆144 and ins214ANRN, and a new VOI N.10 derived from the B.1.1.33 lineage carrying three NTD deletions (∆141-144, ∆211, and ∆256-258). These findings support that the ongoing widespread transmission of SARS-CoV-2 in Brazil generates new viral lineages that might be more resistant to antibody neutralization than parental variants of concern.

The influence of biotinylation on the ability of a computer designed protein to detect B-cells producing anti-HIV-1 2F5 antibodies.

Antibodies against the HIV-1 2F5 epitope are known as one of the most powerful and broadly protective anti-HIV antibodies. Therefore, vaccine strategies that include the 2F5 epitope in their formulation require a robust method to detect specific anti-2F5 antibody production by B cells. Towards this goal, we have biotinylated a previously reported computer-designed protein carrying the HIV-1 2F5 epitope aiming the further development of a platform to detect human B-cells expressing anti-2F5 antibodies through flow cytometry. Biophysical and immunological properties of our devised protein were characterized by computer simulation and experimental methods. Biotinylation did not affect folding and improved protein stability and solubility. The biotinylated protein exhibited similar binding affinity trends compared to its unbiotinylated counterpart and was recognized by anti-HIV-1 2F5 antibodies expressed on the surface of patient-derived peripheral blood mononuclear cells. Moreover, we present a high affinity marker for the identification of epitope-specific B cells that can be used to measure the efficacy of vaccine strategies based on the HIV-1 envelope protein.