ABSTRACT
Vaccination against SARS-CoV-2 with BNT162b2 mRNA vaccine plays a critical role in COVID-19 prevention. Although BNT162b2 is highly effective against COVID-19, a time-dependent decrease in neutralizing antibodies (NAbs) is observed. The aim of this study was to identify the individual features that may predict NAbs levels after vaccination. Machine learning techniques were applied to data from 302 subjects. Principal component analysis (PCA), factor analysis of mixed data (FAMD), k-means clustering, and random forest were used. PCA and FAMD showed that younger subjects had higher levels of neutralizing antibodies than older subjects. The effect of age is strongest near the vaccination date and appears to decrease with time. Obesity was associated with lower antibody response. Gender had no effect on NAbs at nine months, but there was a modest association at earlier time points. Participants with autoimmune disease had lower inhibitory levels than participants without autoimmune disease. K-Means clustering showed the natural grouping of subjects into five categories in which the characteristics of some individuals predominated. Random forest allowed the characteristics to be ordered by importance. Older age, higher body mass index, and the presence of autoimmune diseases had negative effects on the development of NAbs against SARS-CoV-2, nine months after full vaccination.
ABSTRACT
The sustainability of coronavirus 19 (COVID-19) vaccine-induced immunity against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is critical to be determined to inform public health decisions on vaccination programs and prevention measures against COVID-19. The aim of the present study was to prospectively evaluate the kinetics of neutralizing antibodies (NAbs) and anti-S-receptor binding domain (RBD IgGs) against SARS-CoV-2 after full vaccination with the BNT162b2 mRNA vaccine for up to 9 months in healthy individuals (NCT04743388). The assessments were performed at the following time points after the second vaccination: 2 weeks, 1 month, 3 months, 6 months, and 9 months. The measurements were performed with the GenScript's cPassTM SARS-CoV-2 NAbs Detection Kit (GenScript, Inc.; Piscataway, NJ) and the Elecsys Anti-SARS-CoV-2 S assay (Roche Diagnostics GmbH; Mannheim, Germany). Three hundred nine participants with a median age of 48 years were included. A gradual decline in both NAbs and anti-S-RBD IgGs became evident from 2 weeks to 9 months postvaccination. Both NAbs and anti-S-RBD IgGs levels were significantly lower at 9 months compared with the previous timepoints. Interestingly, age was found to exert a statistically significant effect on NAbs elimination only during the first-trimester postvaccination, as older age was associated with a more rapid clearance of NAbs. Furthermore, simulation studies predicted that the median NAb value would fall from 66% at 9 months to 59% and 45% at 12 and 18 months postvaccination, respectively. This finding may reflect a declining degree of immune protection against COVID-19 and advocates for the administration of booster vaccine shots especially in areas with emerging outbreaks.
ABSTRACT
Featured Application Open Information Extraction on the COVID-19 Open Research Dataset (CORD-19). The usefulness of automated information extraction tools in generating structured knowledge from unstructured and semi-structured machine-readable documents is limited by challenges related to the variety and intricacy of the targeted entities, the complex linguistic features of heterogeneous corpora, and the computational availability for readily scaling to large amounts of text. In this paper, we argue that the redundancy and ambiguity of subject-predicate-object (SPO) triples in open information extraction systems has to be treated as an equally important step in order to ensure the quality and preciseness of generated triples. To this end, we propose a pipeline approach for information extraction from large corpora, encompassing a series of natural language processing tasks. Our methodology consists of four steps: i. in-place coreference resolution, ii. extractive text summarization, iii. parallel triple extraction, and iv. entity enrichment and graph representation. We manifest our methodology on a large medical dataset (CORD-19), relying on state-of-the-art tools to fulfil the aforementioned steps and extract triples that are subsequently mapped to a comprehensive ontology of biomedical concepts. We evaluate the effectiveness of our information extraction method by comparing it in terms of precision, recall, and F1-score with state-of-the-art OIE engines and demonstrate its capabilities on a set of data exploration tasks.