A5080649549- Computational Drug Discovery Methods
- Bioinformatics and Genomic Networks
- Biomedical Text Mining and Ontologies
- SARS-CoV-2 and COVID-19 Research
- MicroRNA in disease regulation
- Stroke Rehabilitation and Recovery
- RNA modifications and cancer
- Pharmacogenetics and Drug Metabolism
- Birth, Development, and Health
- Cancer-related molecular mechanisms research
- Spaceflight effects on biology
- Extracellular vesicles in disease
- RNA regulation and disease
- Dementia and Cognitive Impairment Research
- Motor Control and Adaptation
- Autonomous Vehicle Technology and Safety
- Advanced Vision and Imaging
- Microbial Natural Products and Biosynthesis
- Machine Learning in Healthcare
- Long-Term Effects of COVID-19
- Machine Learning in Materials Science
- Erythropoietin and Anemia Treatment
- Neurological Disease Mechanisms and Treatments
- COVID-19 diagnosis using AI
- Receptor Mechanisms and Signaling
Universidad Nacional de Asunción
2021-2024
Fundação Getulio Vargas
2020-2022
Royal Holloway University of London
2018-2020
University of London
2016-2017
Universidad Católica Nuestra Señora de la Asunción
2014-2015
Université Toulouse III - Paul Sabatier
2011
Université de Toulouse
2011
Laboratoire d'Analyse et d'Architecture des Systèmes
2011
Centre National de la Recherche Scientifique
2011
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) viral proteins bind to host mitochondrial proteins, likely inhibiting oxidative phosphorylation (OXPHOS) and stimulating glycolysis. We analyzed gene expression in nasopharyngeal autopsy tissues from patients with disease 2019 (COVID-19). In samples declining titers, the virus blocked transcription of a subset nuclear DNA (nDNA)–encoded OXPHOS genes, induced microRNA 2392, activated HIF-1α induce glycolysis, immune defenses...
Abstract A central issue in drug risk-benefit assessment is identifying frequencies of side effects humans. Currently, are experimentally determined randomised controlled clinical trials. We present a machine learning framework for computationally predicting effects. Our matrix decomposition algorithm learns latent signatures drugs and that both reproducible biologically interpretable. show the usefulness our approach on 759 structurally therapeutically diverse 994 from all human...
Highlights•In silico predictions of miR-2392 as a miRNA involved with SARS-CoV-2•Overexpression produces similar biological response COVID-19 infection•miR-2392 is confirmed to circulate in serum and urine patients COVID-19•Development initiated potential antiviral therapeutic against COVID-19SummaryMicroRNAs (miRNAs) are small non-coding RNAs post-transcriptional gene regulation that have major impact on many diseases provide an exciting avenue toward therapeutics. From patient...
We present two machine learning approaches for drug repurposing. While we have developed them COVID-19, they are disease-agnostic. The methodologies complementary, targeting SARS-CoV-2 and host factors, respectively. Our first approach consists of a matrix factorization algorithm to rank broad-spectrum antivirals. second approach, based on network medicine, uses graph kernels drugs according the perturbation induce subnetwork human interactome that is crucial infection/replication....
Rationale: Viral infections are complex processes based on an intricate network of molecular interactions. The infectious agent hijacks components the cellular machinery for its profit, circumventing natural defense mechanisms triggered by infected cell. successful completion replicative viral cycle within a cell depends function versus defenses. Non-coding RNAs (ncRNAs) important modulators, either promoting or preventing progression infections. Among these ncRNAs, long non-coding RNA...
Abstract Our previous research revealed a key microRNA signature that is associated with spaceflight can be used as biomarker and to develop countermeasure treatments mitigate the damage caused by space radiation. Here, we expand on this work determine biological factors rescued treatment. We performed RNA-sequencing transcriptomic analysis 3D microvessel cell cultures exposed simulated deep radiation (0.5 Gy of Galactic Cosmic Radiation) without antagonists three microRNAs: miR-16-5p,...
Early and accurate detection of side effects is critical for the clinical success drugs under development. Here, we aim to predict unknown with a small number identified in randomized controlled trials. Our machine learning framework, geometric self-expressive model (GSEM), learns globally optimal self-representations from pharmacological graph networks. We show usefulness GSEM on 505 therapeutically diverse 904 multiple human physiological systems. also data integration strategy that could...
The accurate identification of drug side effects represents a major concern for public health. We propose collaborative filtering model large-scale prediction effects. Our approach provides recommendations drugs to safety professionals. proposed latent factor relies solely on the drug-side effect relationships from data. Applied 1,525 marketed and 2,050 terms, we achieved an AUPRC (area under precision- recall curve) 0.342 in test set, with sensitivity 0.73 given specificity 0.95, providing...
MicroRNAs (miRNAs) have been implicated in human disorders, from cancers to infectious diseases. Targeting miRNAs or their target genes with small molecules offers opportunities modulate dysregulated cellular processes linked Yet, predicting associated remains challenging due the size of molecule-miRNA datasets. Herein, we develop a generalized deep learning framework, sChemNET, for affecting miRNA bioactivity based on chemical structure and sequence information. sChemNET overcomes...
Balance control plays a key role in neuromotor rehabilitation after stroke or spinal cord injuries. Computerized dynamic posturography (CDP) is classic technological tool to assess the status of balance and identify potential disorders. Despite more accurate diagnosis generated by these tools, current strategies promote are still limited do not take full advantage technologies available. This paper presents novel training platform which combines CDP device made from low-cost interfaces, such...
Summary MicroRNAs (miRNAs) are small non-coding RNAs involved in post-transcriptional gene regulation that have a major impact on many diseases and provides an exciting avenue towards antiviral therapeutics. From patient transcriptomic data, we discovered circulating miRNA, miR-2392, is directly with SARS-CoV-2 machinery during host infection. Specifically, show miR-2392 key driving downstream suppression of mitochondrial expression, increasing inflammation, glycolysis, hypoxia as well...
This paper presents an embedded architecture to implement in real time the Inverse Perspective Mapping (IPM) algorithm. The IPM algorithm allows a robot detect obstacles under hypothesis that ground is flat, and definition of obstacle being anything has height above ground. based on modifying camera's angle view remove perspective effect for plane. implemented using co-design techniques validated Stratix 3 FPGA. Several approaches proposed develop devoted this algorithm, are compared....
The growing productivity gap between investment in drug research and development (R&D) the number of new medicines approved by US Food Drug Administration (FDA) past decade is concerning. This problem raises need for innovative approaches drug-target prediction a deeper understanding interplay drugs their target proteins. Chemogenomics interdisciplinary field which aims to predict gene/protein/ligand relationships. predictions are based on assumption that chemically similar compounds should...
<title>Abstract</title> In the era of renewed space exploration, comprehending effects environment on human health, particularly for deep missions, is crucial. While extensive research exists impacts spaceflight, there a gap regarding female reproductive risks. We hypothesize that stressors could have enduring potentially increasing risks future pregnancies upon return to Earth, related small-for-gestational-age (SGA) fetuses. To address this, we identify shared microRNA (miRNA) signature...
Abstract Drug side effects are a leading cause of morbidity and mortality. Currently, the frequency drug is determined experimentally during human clinical trials through placebo-controlled studies. Here we present novel framework to computationally predict effects. Our algorithm based on learning latent variable model for drugs by matrix decomposition. Extensive evaluations held out test sets show that class predicted with 67.8% 94% accuracy in neighborhood correct class. Evaluations...
Abstract Pair-input associations for drug-side effects are obtained through expensive placebo-controlled experiments in human clinical trials. An important challenge computational pharmacology is to predict missing given a few entries the effect matrix, as these predictions can be used direct further Here we introduce Geometric Sparse Matrix Completion (GSMC) model predicting drug side effects. Our high-rank matrix completion learns non-negative sparse matrices of coefficients drugs and by...
Drug repositioning is an attractive cost-efficient strategy for the development of treatments human diseases. Here, we propose interpretable model that learns disease self-representations drug repositioning. Our self-representation represents each as a linear combination few other We enforce proximity in learnt representations way to preserve geometric structure phenome network - domain-specific knowledge naturally adds relational inductive bias self-representations. prove our method...
Abstract The identification of missing drug targets is critical for the development treatments and molecular elucidation side effects. Drug have been predicted by exploiting molecular, biological or pharmacological features drugs protein targets. Yet, developing integrative interpretable machine learning models predicting remains a challenging task. We present Inception, an matrix completion model Inception self-expressive that learns two similarity matrices: one another These learned...
The current drug development pipelines are characterised by long processes with high attrition rates and elevated costs. More than 80% of new compounds fail in the later stages testing due to severe side-effects caused unknown biomolecular targets compounds. In this work, we present a measure that can predict shared for drugs DrugBank through large scale analysis biomedical literature. We show using MeSH ontology terms accurately describe appropriate use ontological structure determine...