• Digital Humanities and Cultural Heritage
• Publications close
• Research data close
• Other research products close
• IL close
• Mémoires en Sciences de l'Informati... close

Abstract Background Determining a protein’s quaternary state, i.e. the number of monomers in a functional unit, is a critical step in protein characterization. Many proteins form multimers for their activity, and over 50% are estimated to naturally form homomultimers. Experimental quaternary state determination can be challenging and require extensive work. To complement these efforts, a number of computational tools have been developed for quaternary state prediction, often utilizing experimentally validated structural information. Recently, dramatic advances have been made in the field of deep learning for predicting protein structure and other characteristics. Protein language models, such as ESM-2, that apply computational natural-language models to proteins successfully capture secondary structure, protein cell localization and other characteristics, from a single sequence. Here we hypothesize that information about the protein quaternary state may be contained within protein sequences as well, allowing us to benefit from these novel approaches in the context of quaternary state prediction. Results We generated ESM-2 embeddings for a large dataset of proteins with quaternary state labels from the curated QSbio dataset. We trained a model for quaternary state classification and assessed it on a non-overlapping set of distinct folds (ECOD family level). Our model, named QUEEN (QUaternary state prediction using dEEp learNing), performs worse than approaches that include information from solved crystal structures. However, it successfully learned to distinguish multimers from monomers, and predicts the specific quaternary state with moderate success, better than simple sequence similarity-based annotation transfer. Our results demonstrate that complex, quaternary state related information is included in such embeddings. Conclusions QUEEN is the first to investigate the power of embeddings for the prediction of the quaternary state of proteins. As such, it lays out strengths as well as limitations of a sequence-based protein language model approach, compared to structure-based approaches. Since it does not require any structural information and is fast, we anticipate that it will be of wide use both for in-depth investigation of specific systems, as well as for studies of large sets of protein sequences. A simple colab implementation is available at: https://colab.research.google.com/github/Furman-Lab/QUEEN/blob/main/QUEEN_prediction_notebook.ipynb.

International audience

International audience

The Coxiellaceae family is composed of five genera showing lifestyles ranging from free-living to symbiosis. Among them, Coxiella burnetii is a well-known pathogen causing Q fever in humans. This bacterium presents both intracellular (parasitic) and environmental (resistant) forms. Recently, several environmental Coxiella genomes have been reported, among which several have come from intracellular mutualistic symbionts of ticks, termed Coxiella-like endosymbionts. We sequenced two new Coxiella-LE genomes from Dermacentor marginatus (CLEDm) and Ornithodoros maritimus (CLEOmar) ticks, the latter belonging to the C. burnetii lineage. Using these newly sequenced Coxiella-LEs and 43 Coxiellaceae genomes, we conducted comparative genomic and phylogenomic analyses to increase our knowledge of C. burnetii pathogenicity and the emergence of Coxiella-LEs. Results highlight the probably parasitic nature of the common ancestor of the Coxiellaceae. Indeed, the virulence factor Dot/Icm T4 Secretion System is present in most, but not all, Coxiellaceae. Whereas it is part of a putative pathogenic island in C. burnetii, it has been entirely lost or inactivated in Coxiella-LEs, suggesting its importance in pathogenesis. Additionally, we found that a Sha/Mrp antiporter was laterally acquired in the C. burnetii lineage. This antiporter might be involved in alkali resistance and the development of the resistant form that is able to persist in the environment for long periods of time. The Sha operon is eroded or absent in Coxiella-LEs. Finally, we found that all Coxiella representatives produce B vitamins and co-factors indicating a pre-adaptation of Coxiella to mutualism with hematophagous arthropods. Accordingly, the ancestor of C. burnetii and Coxiella-LEs was likely a parasitic bacterium able to manipulate its host cell and to produce vitamins and co-factors for its own use. International audience

Binaries of relatively massive black holes like GW190521 have been proposed to form in dense gas environments, such as the disks of active galactic nuclei (AGNs), and they might be associated with transient electromagnetic counterparts. The interactions of this putative environment with the binary could leave a significant imprint at the low gravitational wave frequencies observable with the Laser Interferometer Space Antenna (LISA). We show that LISA will be able to detect up to ten GW190521-like black hole binaries, with sky position errors ≲1  deg2. Moreover, it will measure directly various effects due to the orbital motion around the supermassive black hole at the center of the AGN, especially the Doppler modulation and the Shapiro time delay. Thanks to a careful treatment of their frequency domain signal, we were able to perform the full parameter estimation of Doppler and Shapiro-modulated binaries as seen by LISA. We find that the Doppler and Shapiro effects will allow for measuring the AGN parameters (radius and inclination of the orbit around the AGN, central black hole mass) with up to percent-level precision. Properly modeling these low-frequency environmental effects is crucial to determine the binary formation history, as well as to avoid biases in the reconstruction of the source parameters and in tests of general relativity with gravitational waves. International audience

Interglacial periods are characterised by thick accumulations of halite units in the Dead Sea Basin. During these intervals, small water droplets (fluid inclusions, FIs) were entrapped in the halite crystals which serve as windows to estimate the chemistry and physical properties of the primary lake water conditions. Brillouin spectroscopy is used here to reconstruct annual resolution temperatures from a halite core section in the Dead Sea Basin during the onset of Marine Isotope Stage 5e (ca 130 ka) of the Last Interglacial. Lake bottom temperatures can be inferred based on the occurrence of coarse/fine halite facies, as observed today with the formation of equivalent halite facies during winter/summer seasons in the Dead Sea. A recurring increase in lake bottom temperatures is found along the direction of coarse halite layers in three successive years. Moreover, low FI entrapment temperatures were detected in layers of fine (cumulate) halite facies. These results imply a twofold stronger seasonality in the Dead Sea Basin compared to today, with colder winters at the onset of Marine Isotope Stage 5e. The results therefore highlight the potential of using cyclic salt deposits to reconstruct seasonal temperature variability for numerous evaporitic environments in the geological record. International audience

Objectives: Violence against women is a global public health problem. Although there has been much research done on violence against women, there are few studies that provide the current scientific production. Methods: In this study, bibliometric analysis has been used to evaluate the 1984 documents from 1986 to 2020 based on the Scopus database. These documents were analyzed quantitatively by the Bibliometric R Package and the VOS viewer software. In addition, the 20 top-cited papers were analyzed qualitatively. Results: The research findings show that the United States is a leader in this field with the most highly cited articles and also the greatest number of publications followed by the United Kingdom, Canada, Australia, and South Africa. A total of 1984 documents were collected from the Scopus database and were analyzed in the Bibliometric R Research Package and the VOSviewer software. The results demonstrated that the average citations per year for each document were 23.39% and the annual scientific production growth rate was 16.86%. The keywords analysis indicates that most articles focus on “sexual violence”, “sexual assault”, “intimate partner violence”, “violence against women”, “sexual abuse”, “domestic violence”, “child sexual abuse”, “prevention”, and “rape.” Sources such as the “Journal of Interpersonal Violence”, “Journal of Violence Against Woman”, “Journal of Violence and Victims”, “Psychology of Women Quarterly”, “Journal of Adolescent Health”, “Journal of Consulting and Clinical Psychology”, “American Journal of Public Health”, “Journal of Consulting and Clinical Psychology”, and “American Journal of Public Health”, and “The Lancet” are the top most productive in this field. Discussion: Examining the articles showed that the vast majority of women have experienced verbal, sexual, intimate partner violence, cyber harassment, and so on.