The Development of an Ontology Platform to Detect Serine Protease Involvement in COVID-19-Induced Co-Morbid BALT Lymphoma

Mucosal-associated lymphoid tissue (MALT) lymphoma is a rare, low-grade non-Hodgkin lymphoma that develops from B cells and often is diagnosed incidentally. Although it possesses diverse subvariants, such as bronchial-associated lymphoid tissue (BALT), its pathogenesis remains unclear. Currently, little consensus exists on a standard treatment approach partially due to their rare occurrences and heterogeneities, but also due to good prognosis. Studies over the past decades have identified an association between coronaviral infection and BALT formation, whereby mutations in serine proteases increase the risk of BALT lymphoma. The aim of this study was to use a computational approach to validate and further elucidate this relationship specifically between SARS-CoV-2 and BALT lymphoma development.

To address this, ontologies for serine proteolytic homeostasis were constructed utilizing keywords and abstracts from PubMed, which were then used with parsing and logical "OR" functions to generate search inputs for the European Bioinformatics Institute (EBI) COVID-19 ontology and ClinicalTrials.gov (as in Table 1), specifically for and between "BALT", "lymphoma", "serine protease", and "heat shock protein" as a representative type of serine protease inhibitor (SerPIn) class. The Heat Shock Protein 70 family and its more specific subtype Binding immunoglobulin protein (BiP) (a.k.a. GRP-78, HSPA5, and Byun1), were also manually identified as an alternative results marker for efficacy comparison.

Results showed that this approach successfully demonstrated a literary relationship between serine proteases (SERPINE1) with heat shock proteins (HS90) and COVID-19, with aryl hydrocarbon (Row 4 of Table 1) potentially acting as this bridge between BALT lymphomas, transmembrane bound serine proteases, and SARS-CoV-2. This demonstrates a viable approach to automate the development of future ontologies to detail the pathogenesis between SARS-CoV-2, BALT lymphoma, and serine proteases. However, it was limited in its ability to yield positive hit results for HSP70 or GRP-78, likely due to the narrow focus of the EBI COVID-19 ontology and PubMed for covering full text papers. Nonetheless, this platform still offers a wealth of genetic, proteomic, and clinical trials information that can effectively be deployed toward characterizing disease pathogenesis, even in rare hematological malignancies.

Forward looking, future studies will continue to investigate the application of serine proteolytic homeostasis ontologies to BALT lymphoma to better characterize its pathogenesis and interplay with other malignancies, which can then be utilized to develop targeted therapeutic solutions for critical BALT and MALT malignancies.

No relevant conflicts of interest to declare.