scholarly journals Optimising biomedical relationship extraction with BioBERT

2020 ◽  
Author(s):  
Oliver Giles ◽  
Anneli Karlsson ◽  
Spyroula Masiala ◽  
Simon White ◽  
Gianni Cesareni ◽  
...  
Keyword(s):  
Text Mining ◽  
Protein Interactions ◽  
Language Model ◽  
Fine Tuning ◽  
Domain Specific Language ◽  
Protein Protein Interactions ◽  
Domain Specific ◽  
Manual Curation ◽  
Relationship Extraction ◽  
Biological Entities

AbstractText mining is widely used within the life sciences as an evidence stream for inferring relationships between biological entities. In most cases, conventional string matching is used to identify cooccurrences of given entities within sentences. This limits the utility of text mining results, as they tend to contain significant noise due to weak inclusion criteria. We show that, in the indicative case of protein-protein interactions (PPIs), the majority of sentences containing cooccurrences (∽75%) do not describe any causal relationship. We further demonstrate the feasibility of fine tuning a strong domain-specific language model, BioBERT, to analyse sentences containing cooccurrences and accurately (F1 score: 88.95%) identify functional links between proteins. These strong results come in spite of the deep complexity of the language involved, which limits the accuracy even of expert curators. We establish guidelines for best practices in data creation to this end, including an examination of inter-annotator agreement, of semisupervision, and of rules based alternatives to manual curation, and explore the potential for downstream use of the model to accelerate curation of interactions in the SIGNOR database of causal protein interactions and the IntAct database of experimental evidence for physical protein interactions.

scholarly journals Engineering selective competitors for the discrimination of highly conserved protein-protein interaction modules

2019 ◽  
Vol 10 (1) ◽  
Author(s):  
Charlotte Rimbault ◽  
Kashyap Maruthi ◽  
Christelle Breillat ◽  
Camille Genuer ◽  
Sara Crespillo ◽  
...  
Keyword(s):  
Protein Interactions ◽  
Pdz Domain ◽  
Specific Inhibitor ◽  
Rational Approach ◽  
Protein Protein Interactions ◽  
Protein Protein Interaction ◽  
Domain Specific ◽  
Postsynaptic Protein ◽  
Protein Interfaces ◽  
Direct Generation

Abstract Designing highly specific modulators of protein-protein interactions (PPIs) is especially challenging in the context of multiple paralogs and conserved interaction surfaces. In this case, direct generation of selective and competitive inhibitors is hindered by high similarity within the evolutionary-related protein interfaces. We report here a strategy that uses a semi-rational approach to separate the modulator design into two functional parts. We first achieve specificity toward a region outside of the interface by using phage display selection coupled with molecular and cellular validation. Highly selective competition is then generated by appending the more degenerate interaction peptide to contact the target interface. We apply this approach to specifically bind a single PDZ domain within the postsynaptic protein PSD-95 over highly similar PDZ domains in PSD-93, SAP-97 and SAP-102. Our work provides a paralog-selective and domain specific inhibitor of PSD-95, and describes a method to efficiently target other conserved PPI modules.

scholarly journals Latest update on chemokine receptors as therapeutic targets

2021 ◽  
Author(s):  
Wing Yee Lai ◽  
Anja Mueller
Keyword(s):  
Protein Interactions ◽  
Chemokine Receptors ◽  
Inflammatory Diseases ◽  
Specific Binding ◽  
Therapeutic Targets ◽  
Cell Types ◽  
Fine Tuning ◽  
Protein Protein Interactions ◽  
Diverse Range ◽  
Chemokine Ligand

The chemokine system plays a fundamental role in a diverse range of physiological processes, such as homeostasis and immune responses. Dysregulation in the chemokine system has been linked to inflammatory diseases and cancer, which renders chemokine receptors to be considered as therapeutic targets. In the past two decades, around 45 drugs targeting chemokine receptors have been developed, yet only three are clinically approved. The challenging factors include the limited understanding of aberrant chemokine signalling in malignant diseases, high redundancy of the chemokine system, differences between cell types and non-specific binding of the chemokine receptor antagonists due to the broad ligand-binding pockets. In recent years, emerging studies attempt to characterise the chemokine ligand–receptor interactions and the downstream signalling protein–protein interactions, aiming to fine tuning to the promiscuous interplay of the chemokine system for the development of precision medicine. This review will outline the updates on the mechanistic insights in the chemokine system and propose some potential strategies in the future development of targeted therapy.

scholarly journals The marriage of chemokines and galectins as functional heterodimers

2021 ◽  
Author(s):  
Philipp von Hundelshausen ◽  
Kanin Wichapong ◽  
Hans-Joachim Gabius ◽  
Kevin H. Mayo
Keyword(s):  
Protein Interactions ◽  
Role Models ◽  
Cell Types ◽  
Fine Tuning ◽  
Protein Protein Interactions ◽  
Research Activity ◽  
Physiological Processes ◽  
Galectin 3 ◽  
Inflammatory Processes ◽  
Familial Classification

AbstractTrafficking of leukocytes and their local activity profile are of pivotal importance for many (patho)physiological processes. Fittingly, microenvironments are complex by nature, with multiple mediators originating from diverse cell types and playing roles in an intimately regulated manner. To dissect aspects of this complexity, effectors are initially identified and structurally characterized, thus prompting familial classification and establishing foci of research activity. In this regard, chemokines present themselves as role models to illustrate the diversification and fine-tuning of inflammatory processes. This in turn discloses the interplay among chemokines, their cell receptors and cognate glycosaminoglycans, as well as their capacity to engage in new molecular interactions that form hetero-oligomers between themselves and other classes of effector molecules. The growing realization of versatility of adhesion/growth-regulatory galectins that bind to glycans and proteins and their presence at sites of inflammation led to testing the hypothesis that chemokines and galectins can interact with each other by protein–protein interactions. In this review, we present some background on chemokines and galectins, as well as experimental validation of this chemokine–galectin heterodimer concept exemplified with CXCL12 and galectin-3 as proof-of-principle, as well as sketch out some emerging perspectives in this arena.

scholarly journals The STRING database in 2021: customizable protein–protein networks, and functional characterization of user-uploaded gene/measurement sets

2020 ◽  
Vol 49 (D1) ◽  
pp. D605-D612 ◽  
Author(s):  
Damian Szklarczyk ◽  
Annika L Gable ◽  
Katerina C Nastou ◽  
David Lyon ◽  
Rebecca Kirsch ◽  
...  
Keyword(s):  
Text Mining ◽  
Protein Interactions ◽  
Functional Characterization ◽  
Protein Networks ◽  
Protein Protein Interactions ◽  
Genomic Context ◽  
Mining System ◽  
String Database ◽  
System A ◽  
Physical Interactions

Abstract Cellular life depends on a complex web of functional associations between biomolecules. Among these associations, protein–protein interactions are particularly important due to their versatility, specificity and adaptability. The STRING database aims to integrate all known and predicted associations between proteins, including both physical interactions as well as functional associations. To achieve this, STRING collects and scores evidence from a number of sources: (i) automated text mining of the scientific literature, (ii) databases of interaction experiments and annotated complexes/pathways, (iii) computational interaction predictions from co-expression and from conserved genomic context and (iv) systematic transfers of interaction evidence from one organism to another. STRING aims for wide coverage; the upcoming version 11.5 of the resource will contain more than 14 000 organisms. In this update paper, we describe changes to the text-mining system, a new scoring-mode for physical interactions, as well as extensive user interface features for customizing, extending and sharing protein networks. In addition, we describe how to query STRING with genome-wide, experimental data, including the automated detection of enriched functionalities and potential biases in the user's query data. The STRING resource is available online, at https://string-db.org/.

Targeting Kinase Interaction Networks: A New Paradigm in PPI Based Design of Kinase Inhibitors

2019 ◽  
Vol 19 (6) ◽  
pp. 467-485 ◽  
Author(s):  
Pranitha Jenardhanan ◽  
Manivel Panneerselvam ◽  
Premendu P. Mathur
Keyword(s):  
Protein Interactions ◽  
Kinase Inhibitors ◽  
Disease Onset ◽  
Peptide Inhibitors ◽  
Protein Protein Interactions ◽  
New Paradigm ◽  
Diverse Range ◽  
Protein Protein Interaction ◽  
Domain Specific ◽  
And Function

Background: Kinases are key modulators in regulating diverse range of cellular activities and are an essential part of the protein-protein interactome. Understanding the interaction of kinases with different substrates and other proteins is vital to decode the cell signaling machinery as well as causative mechanism for disease onset and progression. Objective: The objective of this review is to present all studies on the structure and function of few important kinases and highlight the protein-protein interaction (PPI) mechanism of kinases and the kinase specific interactome databases and how such studies could be utilized to develop anticancer drugs. Methods: The article is a review of the detailed description of the various domains in kinases that are involved in protein-protein interactions and specific inhibitors developed targeting these PPI domains. Results: The review has surfaced in depth the interacting domains in key kinases and their features and the roles of PPI in the human kinome and the various signaling cascades that are involved in certain types of cancer. Conclusion: The insight availed into the mechanism of existing peptide inhibitors and peptidomimetics against kinases will pave way for the design and generation of domain specific peptide inhibitors with better productivity and efficiency and the various software and servers available can be of great use for the identification and analysis of protein-protein interactions.

Sign in / Sign up

Export Citation Format

Share Document