Objective: The objective of this diagnostic accuracy review is to evaluate the effectiveness of rapid antigen tests versus viral genetic PCR-based tests on COVID-19 diagnostic accuracy in adults 18 years and over. Introduction: Due to the rapidly changing nature of the COVID-19 pandemic, it is imperative that clinicians have access to the most relevant and effective tools and information required to combat this disease. Testing strategies are being developed continuously and need to be evaluated to ensure their appropriate implementation into clinical practice. Inclusion criteria: This systematic review will include publications that are in the English language (originally or translated) and any gray literature pertaining to the tests of interest. All races, ages over 18, and geographic locations will be considered. Methods: MEDLINE (PubMed), Embase (Elsevier), Scopus (Elsevier), Qinsight (Quertle), and WHO COVID-19 database (World Health Organization) will be searched. Scopus, Qinsight, and WHO COVID-19 include gray literature. Studies in English published from November 2019 to the present will be considered. Animal studies and studies including pregnant women will be excluded. Retrieval of full-text studies, data extraction, and assessment of methodological quality will be performed independently by two reviewers. A custom data extraction table will be used. Findings will be graphically represented with two forest plots, one for sensitivity and the other for specificity. The strategy for meta-analysis includes producing a summary receiver operating characteristic curve and estimating the summary sensitivity/specificity for each threshold provided in the articles. Systematic review registration number: PROSPERO CRD42020224250

RIG-I-like receptors (RLRs) are DEAD-box helicases and cytoplasmic pathogen recognition receptors that recognize and bind to nonself/viral RNA and trigger innate antiviral immunity. RLRs are essential for the detection of RNA virus infection and include RIG-I, MDA5, and LGP2. LGP2 has been implicated as RIG-I or MDA5 co-factor to regulate their activity. We conducted a yeast 2-hybrid (Y2H) screen (using LGP2 as bait) of a human hepatic cDNA library to identify protein binding partners of LGP2. Identified proteins were validated as LGP2 interactors through overexpression and co-immunoprecipitation assays of LGP2 binding. DDX39A, a DEAD Box helicase protein with known roles in RNA splicing, was identified as a binding partner. DDX39A overexpression was found to inhibit interferon-beta production from RLR-signaling in a cell-based ligand-free model in a dose-dependent manner. Further, in a cell-based interferon beta-promoter-luciferase assay model of Sendai virus (SenV) infection, DDX39A and LGP2 synergistically inhibited promoter activity. Similar results were seen in a cell-based NFkB-luciferase assay of the same system. Endogenous expression of DDX39A was assayed through the Huh7 human hepatoma cell line system with various stimuli including viral infection with SenV and Interferon-beta treatment showing that DDX39A is constitutively expressed in those cells. Structure-function mutants of DDX39A and LGP2 were used to determine the protein domains and functions required for interaction via co-immunoprecipitation assays. Through these studies, we will define the interactions between LGP2 and DDX39A, and reveal the roles these proteins have as co-factors in RLR signaling.