LIVIVO - The Search Portal for Life Sciences

zur deutschen Oberfläche wechseln
Advanced search

Search results

Result 1 - 10 of total 65

Search options

  1. Article ; Online: Extracting Vocal Biomarkers for Pulmonary Congestion With a Smartphone App.

    Ravindra, Neal G / Kao, David P

    JACC. Heart failure

    2021  Volume 10, Issue 1, Page(s) 50–51

    MeSH term(s) Biomarkers ; Heart Failure ; Humans ; Mobile Applications ; Pulmonary Edema/diagnostic imaging ; Pulmonary Edema/etiology ; Smartphone ; Speech
    Chemical Substances Biomarkers
    Language English
    Publishing date 2021-12-08
    Publishing country United States
    Document type Editorial ; Research Support, N.I.H., Extramural ; Comment
    ZDB-ID 2705621-1
    ISSN 2213-1787 ; 2213-1779
    ISSN (online) 2213-1787
    ISSN 2213-1779
    DOI 10.1016/j.jchf.2021.10.007
    Database MEDical Literature Analysis and Retrieval System OnLINE

    More links

    Kategorien

  2. Article ; Online: Patient-centered and Patient-independent Technologies in Acute Neurological Injury and the Art of Making Useful Medical Contributions: 
An Interview with Kevin Sheth, MD.

    Ravindra, Neal G

    The Yale journal of biology and medicine

    2018  Volume 91, Issue 3, Page(s) 345–351

    MeSH term(s) Brain Ischemia/diagnosis ; Entrepreneurship ; Humans ; Neurosciences/methods ; Physician-Patient Relations ; Physicians ; Stroke/diagnosis
    Language English
    Publishing date 2018-09-21
    Publishing country United States
    Document type Interview
    ZDB-ID 200515-3
    ISSN 1551-4056 ; 0044-0086
    ISSN (online) 1551-4056
    ISSN 0044-0086
    Database MEDical Literature Analysis and Retrieval System OnLINE

    More links

    Kategorien

  3. Thesis ; Online: Molecular Assembly in the Endocytic Pathway

    Ravindra, Neal G.

    2019  

    Abstract: Proteins assembled into cellular pathways often possess non-catalytic, protein-interaction domains. Src-homology 3 (SH3) domains are protein-interaction domains that spatiotemporally connect molecules through transient binding interactions, recognizing ... ...

    Abstract Proteins assembled into cellular pathways often possess non-catalytic, protein-interaction domains. Src-homology 3 (SH3) domains are protein-interaction domains that spatiotemporally connect molecules through transient binding interactions, recognizing linear peptide motifs and localizing proteins to various sub-cellular structures. In the endocytic pathway, there are many SH3-domain-containing proteins and several endocytic proteins contain multiple SH3 domains. I sought to interrogate the degeneracy in the number of SH3 domains within endocytosis and within endocytic proteins and to clarify the influence of each SH3 domain on the assembly and dynamics of the endocytic molecular machinery. To this end, in collaboration with Ronan Fernandez, I created a comprehensive library of endogenous, single SH3 domain deletions in the fission yeast Schizosaccharomyces pombe and used quantitative fluorescence microscopy to measure the effects of these deletions in vivo. I found that endocytic SH3 domains restrict, enhance, or have minor or redundant effects on actin assembly in endocytosis. I also found that some SH3 domains influence the cell’s ability to regulate the number of endocytic events. These observations are consistent with simulated perturbations to reaction steps in the Arp2/3 activation pathway, supporting the explanation that SH3 domains are regulators of Arp2/3-mediated actin nucleation in endocytosis. To investigate the endocytic localization dependence of SH3-domain containing proteins on their SH3 domain(s), in collaboration with Ronan Fernandez, we created a library of single SH3 domain deletions within strains where each SH3 domain’s native protein was also tagged with a fluorescent reporter. Analysis of the localization of these proteins and their fluorescent distribution in live cells reveals that most SH3 domains influence their protein’s localization and assembly dynamics into endocytic structures. Furthermore, several SH3 domains are required for robust localization of their protein to endocytic structures while being dispensable for their protein’s expression. Thus, endocytic SH3 domains may influence the assembly dynamics of SH3-domain-containing proteins into endocytic structures in addition to playing other assembly and regulatory roles within endocytic structures. Given that SH3 domains participate in a large number of interactions in the endocytic protein-interaction network, relative to other modular domains, a plausible answer to how endocytic proteins are recruited may be through SH3 domain-mediated interactions. Yet, one challenge to the use of SH3 domains in synthetic biology is that it is poorly understood how distinct sets of SH3 domains interact with distinct sets of proteins, given the potential overlap between SH3 domain-mediated interactions. To address how SH3 domains assemble proteins into distinct pathways, I proposed that SH3 domains achieve binding specificity through domain-mediated specificity, where binding preferences emerge from unique biophysical properties, and/or through contextual specificity, where binding preferences emerge through unique molecular and cellular environments. I hypothesized that SH3 domains primarily exhibit contextual specificity, which implies that individual SH3 domains are interchangeable. To determine the interchangeability of SH3 domains in a single context, I replaced native endocytic SH3 domains with non-native SH3 domains from other proteins and organisms. Contrary to my suppositions, my findings support the hypothesis that SH3 domains achieve interaction specificity primarily through domain-mediated specificity. However, my results do not entirely rule out contextually-mediated interaction specificity. Collectively, I describe a range of influences and activities that individual SH3 domains have on molecular assembly during endocytosis. The quantitative measurements of molecular assembly during endocytosis described in this dissertation, especially in the background of single deletions of each SH3 domain in endocytosis, reveal that SH3 domains have a variety of influences on actin assembly, endocytosis and the cell’s regulation of the endocytic rate. In particular, SH3 domains appear to play assembly and regulatory roles during endocytosis, perhaps by mediating interactions in the Arp2/3 activation pathway and by influencing the assembly dynamics of SH3 domain-containing proteins and actin accessory factors in the cell. These results add nuance to the purported role of SH3 domains in inducing phase-separated structures that promote local actin assembly in the cell. By providing precise quantitative descriptions into molecular assembly during endocytosis under a variety of perturbations to SH3 domains, this dissertation may inform future synthetic manipulations of endocytosis, especially by deleting or inserting SH3 domains as interchangeable parts in molecular circuits to predictably modulate the activity of the endocytic pathway and govern biological processes relevant to human health.
    Keywords Biophysics|Cellular biology|Molecular biology
    Subject code 572
    Language ENG
    Publishing date 2019-01-01 00:00:01.0
    Publisher Yale University
    Publishing country us
    Document type Thesis ; Online
    Database BASE - Bielefeld Academic Search Engine (life sciences selection)

    More links

    Kategorien

  4. Article: Gaining insight into SARS-CoV-2 infection and COVID-19 severity using self-supervised edge features and Graph Neural Networks

    Sehanobish, Arijit / Ravindra, Neal G. / Dijk, David van

    Abstract: Graph Neural Networks (GNN) have been extensively used to extract meaningful representations from graph structured data and to perform predictive tasks such as node classification and link prediction. In recent years, there has been a lot of work ... ...

    Abstract Graph Neural Networks (GNN) have been extensively used to extract meaningful representations from graph structured data and to perform predictive tasks such as node classification and link prediction. In recent years, there has been a lot of work incorporating edge features along with node features for prediction tasks. In this work, we present a framework for creating new edge features, via a combination of self-supervised and unsupervised learning which we then use along with node features for node classification tasks. We validate our work on two biological datasets comprising of single-cell RNA sequencing data of \textit{in vitro} SARS-CoV-2 infection and human COVID-19 patients. We demonstrate that our method achieves better performance over baseline Graph Attention Network (GAT) and Graph Convolutional Network (GCN) models. Furthermore, given the attention mechanism on edge and node features, we are able to interpret the cell types and genes that determine the course and severity of COVID-19, contributing to a growing list of potential disease biomarkers and therapeutic targets.
    Keywords covid19
    Publisher ArXiv
    Document type Article
    Database COVID19

    Kategorien

  5. Article: Self-supervised edge features for improved Graph Neural Network training

    Sehanobish, Arijit / Ravindra, Neal G. / Dijk, David van

    Abstract: Graph Neural Networks (GNN) have been extensively used to extract meaningful representations from graph structured data and to perform predictive tasks such as node classification and link prediction. In recent years, there has been a lot of work ... ...

    Abstract Graph Neural Networks (GNN) have been extensively used to extract meaningful representations from graph structured data and to perform predictive tasks such as node classification and link prediction. In recent years, there has been a lot of work incorporating edge features along with node features for prediction tasks. One of the main difficulties in using edge features is that they are often handcrafted, hard to get, specific to a particular domain, and may contain redundant information. In this work, we present a framework for creating new edge features, applicable to any domain, via a combination of self-supervised and unsupervised learning. In addition to this, we use Forman-Ricci curvature as an additional edge feature to encapsulate the local geometry of the graph. We then encode our edge features via a Set Transformer and combine them with node features extracted from popular GNN architectures for node classification in an end-to-end training scheme. We validate our work on three biological datasets comprising of single-cell RNA sequencing data of neurological disease, \textit{in vitro} SARS-CoV-2 infection, and human COVID-19 patients. We demonstrate that our method achieves better performance on node classification tasks over baseline Graph Attention Network (GAT) and Graph Convolutional Network (GCN) models. Furthermore, given the attention mechanism on edge and node features, we are able to interpret the cell types and genes that determine the course and severity of COVID-19, contributing to a growing list of potential disease biomarkers and therapeutic targets.
    Keywords covid19
    Publisher ArXiv
    Document type Article
    Database COVID19

    Kategorien

  6. Book ; Online: Gaining Insight into SARS-CoV-2 Infection and COVID-19 Severity Using Self-supervised Edge Features and Graph Neural Networks

    Sehanobish, Arijit / Ravindra, Neal G. / van Dijk, David

    2020  

    Abstract: A molecular and cellular understanding of how SARS-CoV-2 variably infects and causes severe COVID-19 remains a bottleneck in developing interventions to end the pandemic. We sought to use deep learning to study the biology of SARS-CoV-2 infection and ... ...

    Abstract A molecular and cellular understanding of how SARS-CoV-2 variably infects and causes severe COVID-19 remains a bottleneck in developing interventions to end the pandemic. We sought to use deep learning to study the biology of SARS-CoV-2 infection and COVID-19 severity by identifying transcriptomic patterns and cell types associated with SARS-CoV-2 infection and COVID-19 severity. To do this, we developed a new approach to generating self-supervised edge features. We propose a model that builds on Graph Attention Networks (GAT), creates edge features using self-supervised learning, and ingests these edge features via a Set Transformer. This model achieves significant improvements in predicting the disease state of individual cells, given their transcriptome. We apply our model to single-cell RNA sequencing datasets of SARS-CoV-2 infected lung organoids and bronchoalveolar lavage fluid samples of patients with COVID-19, achieving state-of-the-art performance on both datasets with our model. We then borrow from the field of explainable AI (XAI) to identify the features (genes) and cell types that discriminate bystander vs. infected cells across time and moderate vs. severe COVID-19 disease. To the best of our knowledge, this represents the first application of deep learning to identifying the molecular and cellular determinants of SARS-CoV-2 infection and COVID-19 severity using single-cell omics data.

    Comment: To appear at AAAI'21. Previous version (v2) accepted as a spotlight talk at ICML 2020 Workshop on Graph Representation Learning and Beyond (GRL+) and recipient of best paper award for Covid-19 applications. Significant improvements over v2
    Keywords Computer Science - Machine Learning ; Quantitative Biology - Genomics ; Statistics - Machine Learning
    Subject code 006
    Publishing date 2020-06-23
    Publishing country us
    Document type Book ; Online
    Database BASE - Bielefeld Academic Search Engine (life sciences selection)

    More links

    Kategorien

  7. Book ; Online: Self-supervised edge features for improved Graph Neural Network training

    Sehanobish, Arijit / Ravindra, Neal G. / van Dijk, David

    2020  

    Abstract: Graph Neural Networks (GNN) have been extensively used to extract meaningful representations from graph structured data and to perform predictive tasks such as node classification and link prediction. In recent years, there has been a lot of work ... ...

    Abstract Graph Neural Networks (GNN) have been extensively used to extract meaningful representations from graph structured data and to perform predictive tasks such as node classification and link prediction. In recent years, there has been a lot of work incorporating edge features along with node features for prediction tasks. One of the main difficulties in using edge features is that they are often handcrafted, hard to get, specific to a particular domain, and may contain redundant information. In this work, we present a framework for creating new edge features, applicable to any domain, via a combination of self-supervised and unsupervised learning. In addition to this, we use Forman-Ricci curvature as an additional edge feature to encapsulate the local geometry of the graph. We then encode our edge features via a Set Transformer and combine them with node features extracted from popular GNN architectures for node classification in an end-to-end training scheme. We validate our work on three biological datasets comprising of single-cell RNA sequencing data of neurological disease, \textit{in vitro} SARS-CoV-2 infection, and human COVID-19 patients. We demonstrate that our method achieves better performance on node classification tasks over baseline Graph Attention Network (GAT) and Graph Convolutional Network (GCN) models. Furthermore, given the attention mechanism on edge and node features, we are able to interpret the cell types and genes that determine the course and severity of COVID-19, contributing to a growing list of potential disease biomarkers and therapeutic targets.

    Comment: Comments welcome. arXiv admin note: substantial text overlap with arXiv:2006.12971
    Keywords Electrical Engineering and Systems Science - Image and Video Processing ; Computer Science - Machine Learning ; Quantitative Biology - Genomics ; Statistics - Machine Learning ; I.2.4 ; J.3 ; covid19
    Subject code 006
    Publishing date 2020-06-23
    Publishing country us
    Document type Book ; Online
    Database BASE - Bielefeld Academic Search Engine (life sciences selection)

    More links

    Kategorien

  8. Article ; Online: Molecular mechanisms of force production in clathrin-mediated endocytosis.

    Lacy, Michael M / Ma, Rui / Ravindra, Neal G / Berro, Julien

    FEBS letters

    2018  Volume 592, Issue 21, Page(s) 3586–3605

    Abstract: During clathrin-mediated endocytosis (CME), a flat patch of membrane is invaginated and pinched off to release a vesicle into the cytoplasm. In yeast CME, over 60 proteins-including a dynamic actin meshwork-self-assemble to deform the plasma membrane. ... ...

    Abstract During clathrin-mediated endocytosis (CME), a flat patch of membrane is invaginated and pinched off to release a vesicle into the cytoplasm. In yeast CME, over 60 proteins-including a dynamic actin meshwork-self-assemble to deform the plasma membrane. Several models have been proposed for how actin and other molecules produce the forces necessary to overcome the mechanical barriers of membrane tension and turgor pressure, but the precise mechanisms and a full picture of their interplay are still not clear. In this review, we discuss the evidence for these force production models from a quantitative perspective and propose future directions for experimental and theoretical work that could clarify their various contributions.
    MeSH term(s) Actin Cytoskeleton/metabolism ; Actins/metabolism ; Cell Membrane/metabolism ; Clathrin/metabolism ; Endocytosis ; Fungal Proteins/metabolism ; Models, Biological ; Saccharomyces cerevisiae/metabolism ; Schizosaccharomyces/metabolism
    Chemical Substances Actins ; Clathrin ; Fungal Proteins
    Language English
    Publishing date 2018-07-28
    Publishing country England
    Document type Journal Article ; Research Support, N.I.H., Extramural ; Research Support, Non-U.S. Gov't ; Review
    ZDB-ID 212746-5
    ISSN 1873-3468 ; 0014-5793
    ISSN (online) 1873-3468
    ISSN 0014-5793
    DOI 10.1002/1873-3468.13192
    Database MEDical Literature Analysis and Retrieval System OnLINE

    More links

    Kategorien

  9. Article ; Online: Applications of artificial intelligence and machine learning in heart failure.

    Averbuch, Tauben / Sullivan, Kristen / Sauer, Andrew / Mamas, Mamas A / Voors, Adriaan A / Gale, Chris P / Metra, Marco / Ravindra, Neal / Van Spall, Harriette G C

    European heart journal. Digital health

    2022  Volume 3, Issue 2, Page(s) 311–322

    Abstract: Machine learning (ML) is a sub-field of artificial intelligence that uses computer algorithms to extract patterns from raw data, acquire knowledge without human input, and apply this knowledge for various tasks. Traditional statistical methods that ... ...

    Abstract Machine learning (ML) is a sub-field of artificial intelligence that uses computer algorithms to extract patterns from raw data, acquire knowledge without human input, and apply this knowledge for various tasks. Traditional statistical methods that classify or regress data have limited capacity to handle large datasets that have a low signal-to-noise ratio. In contrast to traditional models, ML relies on fewer assumptions, can handle larger and more complex datasets, and does not require predictors or interactions to be pre-specified, allowing for novel relationships to be detected. In this review, we discuss the rationale for the use and applications of ML in heart failure, including disease classification, early diagnosis, early detection of decompensation, risk stratification, optimal titration of medical therapy, effective patient selection for devices, and clinical trial recruitment. We discuss how ML can be used to expedite implementation and close healthcare gaps in learning healthcare systems. We review the limitations of ML, including opaque logic and unreliable model performance in the setting of data errors or data shift. Whilst ML has great potential to improve clinical care and research in HF, the applications must be externally validated in prospective studies for broad uptake to occur.
    Language English
    Publishing date 2022-05-13
    Publishing country England
    Document type Journal Article ; Review
    ISSN 2634-3916
    ISSN (online) 2634-3916
    DOI 10.1093/ehjdh/ztac025
    Database MEDical Literature Analysis and Retrieval System OnLINE

    More links

    Kategorien

  10. Article ; Online: Prediction of neuropathologic lesions from clinical data.

    Phongpreecha, Thanaphong / Cholerton, Brenna / Bukhari, Syed / Chang, Alan L / De Francesco, Davide / Thuraiappah, Melan / Godrich, Dana / Perna, Amalia / Becker, Martin G / Ravindra, Neal G / Espinosa, Camilo / Kim, Yeasul / Berson, Eloise / Mataraso, Samson / Sha, Sharon J / Fox, Edward J / Montine, Kathleen S / Baker, Laura D / Craft, Suzanne /
    White, Lon / Poston, Kathleen L / Beecham, Gary / Aghaeepour, Nima / Montine, Thomas J

    Alzheimer's & dementia : the journal of the Alzheimer's Association

    2023  Volume 19, Issue 7, Page(s) 3005–3018

    Abstract: Introduction: Post-mortem analysis provides definitive diagnoses of neurodegenerative diseases; however, only a few can be diagnosed during life.: Methods: This study employed statistical tools and machine learning to predict 17 neuropathologic ... ...

    Abstract Introduction: Post-mortem analysis provides definitive diagnoses of neurodegenerative diseases; however, only a few can be diagnosed during life.
    Methods: This study employed statistical tools and machine learning to predict 17 neuropathologic lesions from a cohort of 6518 individuals using 381 clinical features (Table S1). The multisite data allowed validation of the model's robustness by splitting train/test sets by clinical sites. A similar study was performed for predicting Alzheimer's disease (AD) neuropathologic change without specific comorbidities.
    Results: Prediction results show high performance for certain lesions that match or exceed that of research annotation. Neurodegenerative comorbidities in addition to AD neuropathologic change resulted in compounded, but disproportionate, effects across cognitive domains as the comorbidity number increased.
    Discussion: Certain clinical features could be strongly associated with multiple neurodegenerative diseases, others were lesion-specific, and some were divergent between lesions. Our approach could benefit clinical research, and genetic and biomarker research by enriching cohorts for desired lesions.
    MeSH term(s) Humans ; Alzheimer Disease/pathology ; Comorbidity ; Neuropathology ; Biomarkers
    Chemical Substances Biomarkers
    Language English
    Publishing date 2023-01-21
    Publishing country United States
    Document type Journal Article ; Research Support, N.I.H., Extramural
    ZDB-ID 2211627-8
    ISSN 1552-5279 ; 1552-5260
    ISSN (online) 1552-5279
    ISSN 1552-5260
    DOI 10.1002/alz.12921
    Database MEDical Literature Analysis and Retrieval System OnLINE

    More links

    Kategorien

To top