Ensemble of deep learning language models to support the creation of living systematic reviews for the COVID-19 literature.

BACKGROUND: The COVID-19 pandemic has led to an unprecedented amount of scientific publications, growing at a pace never seen before. Multiple living systematic reviews have been developed to assist professionals with up-to-date and trustworthy health information, but it is increasingly challenging for systematic reviewers to keep up with the evidence in electronic databases. We aimed to investigate deep learning-based machine learning algorithms to classify COVID-19-related publications to help scale up the epidemiological curation process. METHODS: In this retrospective study, five different pre-trained deep learning-based language models were fine-tuned on a dataset of 6365 publications manually classified into two classes, three subclasses, and 22 sub-subclasses relevant for epidemiological triage purposes. In a k-fold cross-validation setting, each standalone model was assessed on a classification task and compared against an ensemble, which takes the standalone model predictions as input and uses different strategies to infer the optimal article class. A ranking task was also considered, in which the model outputs a ranked list of sub-subclasses associated with the article. RESULTS: The ensemble model significantly outperformed the standalone classifiers, achieving a F1-score of 89.2 at the class level of the classification task. The difference between the standalone and ensemble models increases at the sub-subclass level, where the ensemble reaches a micro F1-score of 70% against 67% for the best-performing standalone model. For the ranking task, the ensemble obtained the highest recall@3, with a performance of 89%. Using an unanimity voting rule, the ensemble can provide predictions with higher confidence on a subset of the data, achieving detection of original papers with a F1-score up to 97% on a subset of 80% of the collection instead of 93% on the whole dataset. CONCLUSION: This study shows the potential of using deep learning language models to perform triage of COVID-19 references efficiently and support epidemiological curation and review. The ensemble consistently and significantly outperforms any standalone model. Fine-tuning the voting strategy thresholds is an interesting alternative to annotate a subset with higher predictive confidence.

Siglec-7 represents a glyco-immune checkpoint for non-exhausted effector memory CD8+ T cells with high functional and metabolic capacities.

While inhibitory Siglec receptors are known to regulate myeloid cells, less is known about their expression and function in lymphocytes subsets. Here we identified Siglec-7 as a glyco-immune checkpoint expressed on non-exhausted effector memory CD8+ T cells that exhibit high functional and metabolic capacities. Seahorse analysis revealed higher basal respiration and glycolysis levels of Siglec-7+ CD8+ T cells in steady state, and particularly upon activation. Siglec-7 polarization into the T cell immune synapse was dependent on sialoglycan interactions in trans and prevented actin polarization and effective T cell responses. Siglec-7 ligands were found to be expressed on both leukemic stem cells and acute myeloid leukemia (AML) cells suggesting the occurrence of glyco-immune checkpoints for Siglec-7+ CD8+ T cells, which were found in patients' peripheral blood and bone marrow. Our findings project Siglec-7 as a glyco-immune checkpoint and therapeutic target for T cell-driven disorders and cancer.

Reducing systematic review burden using Deduklick: a novel, automated, reliable, and explainable deduplication algorithm to foster medical research.

BACKGROUND: Identifying and removing reference duplicates when conducting systematic reviews (SRs) remain a major, time-consuming issue for authors who manually check for duplicates using built-in features in citation managers. To address issues related to manual deduplication, we developed an automated, efficient, and rapid artificial intelligence-based algorithm named Deduklick. Deduklick combines natural language processing algorithms with a set of rules created by expert information specialists. METHODS: Deduklick's deduplication uses a multistep algorithm of data normalization, calculates a similarity score, and identifies unique and duplicate references based on metadata fields, such as title, authors, journal, DOI, year, issue, volume, and page number range. We measured and compared Deduklick's capacity to accurately detect duplicates with the information specialists' standard, manual duplicate removal process using EndNote on eight existing heterogeneous datasets. Using a sensitivity analysis, we manually cross-compared the efficiency and noise of both methods. DISCUSSION: Deduklick achieved average recall of 99.51%, average precision of 100.00%, and average F1 score of 99.75%. In contrast, the manual deduplication process achieved average recall of 88.65%, average precision of 99.95%, and average F1 score of 91.98%. Deduklick achieved equal to higher expert-level performance on duplicate removal. It also preserved high metadata quality and drastically reduced time spent on analysis. Deduklick represents an efficient, transparent, ergonomic, and time-saving solution for identifying and removing duplicates in SRs searches. Deduklick could therefore simplify SRs production and represent important advantages for scientists, including saving time, increasing accuracy, reducing costs, and contributing to quality SRs.

Vaccine Development in the Time of COVID-19: The Relevance of the Risklick AI to Assist in Risk Assessment and Optimize Performance.

The 2019 coronavirus (COVID-19) pandemic revealed the urgent need for the acceleration of vaccine development worldwide. Rapid vaccine development poses numerous risks for each category of vaccine technology. By using the Risklick artificial intelligence (AI), we estimated the risks associated with all types of COVID-19 vaccine during the early phase of vaccine development. We then performed a postmortem analysis of the probability and the impact matrix calculations by comparing the 2020 prognosis to the contemporary situation. We used the Risklick AI to evaluate the risks and their incidence associated with vaccine development in the early stage of the COVID-19 pandemic. Our analysis revealed the diversity of risks among vaccine technologies currently used by pharmaceutical companies providing vaccines. This analysis highlighted the current and future potential pitfalls connected to vaccine production during the COVID-19 pandemic. Hence, the Risklick AI appears as an essential tool in vaccine development for the treatment of COVID-19 in order to formally anticipate the risks, and increases the overall performance from the production to the distribution of the vaccines. The Risklick AI could, therefore, be extended to other fields of research and development and represent a novel opportunity in the calculation of production-associated risks.

Pulmonary mesenchymal stem cells are engaged in distinct steps of host response to respiratory syncytial virus infection.

Lung-resident (LR) mesenchymal stem and stromal cells (MSCs) are key elements of the alveolar niche and fundamental regulators of homeostasis and regeneration. We interrogated their function during virus-induced lung injury using the highly prevalent respiratory syncytial virus (RSV) which causes severe outcomes in infants. We applied complementary approaches with primary pediatric LR-MSCs and a state-of-the-art model of human RSV infection in lamb. Remarkably, RSV-infection of pediatric LR-MSCs led to a robust activation, characterized by a strong antiviral and pro-inflammatory phenotype combined with mediators related to T cell function. In line with this, following in vivo infection, RSV invades and activates LR-MSCs, resulting in the expansion of the pulmonary MSC pool. Moreover, the global transcriptional response of LR-MSCs appears to follow RSV disease, switching from an early antiviral signature to repair mechanisms including differentiation, tissue remodeling, and angiogenesis. These findings demonstrate the involvement of LR-MSCs during virus-mediated acute lung injury and may have therapeutic implications.

Utilizing Artificial Intelligence to Manage COVID-19 Scientific Evidence Torrent with Risklick AI: A Critical Tool for Pharmacology and Therapy Development.

INTRODUCTION: The SARS-CoV-2 pandemic has led to one of the most critical and boundless waves of publications in the history of modern science. The necessity to find and pursue relevant information and quantify its quality is broadly acknowledged. Modern information retrieval techniques combined with artificial intelligence (AI) appear as one of the key strategies for COVID-19 living evidence management. Nevertheless, most AI projects that retrieve COVID-19 literature still require manual tasks. METHODS: In this context, we pre-sent a novel, automated search platform, called Risklick AI, which aims to automatically gather COVID-19 scientific evidence and enables scientists, policy makers, and healthcare professionals to find the most relevant information tailored to their question of interest in real time. RESULTS: Here, we compare the capacity of Risklick AI to find COVID-19-related clinical trials and scientific publications in comparison with clinicaltrials.gov and PubMed in the field of pharmacology and clinical intervention. DISCUSSION: The results demonstrate that Risklick AI is able to find COVID-19 references more effectively, both in terms of precision and recall, compared to the baseline platforms. Hence, Risklick AI could become a useful alternative assistant to scientists fighting the COVID-19 pandemic.

The Distinct Roles of Sialyltransferases in Cancer Biology and Onco-Immunology.

Aberrant glycosylation is a key feature of malignant transformation. Hypersialylation, the enhanced expression of sialic acid-terminated glycoconjugates on the cell surface, has been linked to immune evasion and metastatic spread, eventually by interaction with sialoglycan-binding lectins, including Siglecs and selectins. The biosynthesis of tumor-associated sialoglycans involves sialyltransferases, which are differentially expressed in cancer cells. In this review article, we provide an overview of the twenty human sialyltransferases and their roles in cancer biology and immunity. A better understanding of the individual contribution of select sialyltransferases to the tumor sialome may lead to more personalized strategies for the treatment of cancer.

Siglec-9 Regulates an Effector Memory CD8+ T-cell Subset That Congregates in the Melanoma Tumor Microenvironment.

Emerging evidence suggests an immunosuppressive role of altered tumor glycosylation due to downregulation of innate immune responses via immunoregulatory Siglecs. In contrast, human T cells, a major anticancer effector cell, only rarely express Siglecs. However, here, we report that the majority of intratumoral, but not peripheral blood, cytotoxic CD8+ T cells expressed Siglec-9 in melanoma. We identified Siglec-9+ CD8+ T cells as a subset of effector memory cells with high functional capacity and signatures of clonal expansion. This cytotoxic T-cell subset was functionally inhibited in the presence of Siglec-9 ligands or by Siglec-9 engagement by specific antibodies. TCR signaling pathways and key effector functions (cytotoxicity, cytokine production) of CD8+ T cells were suppressed by Siglec-9 engagement, which was associated with the phosphorylation of the inhibitory protein tyrosine phosphatase SHP-1, but not SHP-2. Expression of cognate Siglec-9 ligands was observed on the majority of tumor cells in primary and metastatic melanoma specimens. Targeting the tumor-restricted, glycosylation-dependent Siglec-9 axis may unleash this intratumoral T-cell subset, while confining T-cell activation to the tumor microenvironment.

A Cartography of Siglecs and Sialyltransferases in Gynecologic Malignancies: Is There a Road Towards a Sweet Future?

Altered surface glycosylation is a key feature of cancers, including gynecologic malignancies. Hypersialylation, the overexpression of sialic acid, is known to promote tumor progression and to dampen antitumor responses by mechanisms that also involve sialic acid binding immunoglobulin-like lectins (Siglecs), inhibitory immune receptors. Here, we discuss the expression patterns of Siglecs and sialyltransferases (STs) in gynecologic cancers, including breast, ovarian, and uterine malignancies, based on evidence from The Cancer Genome Atlas. The balance between sialosides generated by specific STs within the tumor microenvironment and Siglecs on leukocytes may play a decisive role for antitumor immunity. An interdisciplinary effort is required to decipher the characteristics and biological impact of the altered tumor sialome in gynecologic cancers and to exploit this knowledge to the clinical benefit of patients.