SARS-CoV-2 Mutation Study Reveals Data Processing Errors in Genetic Databases

A study by the Institute for Integrative Systems Biology (CSIC) and the University of Valencia has revealed data processing errors in the GISAID database, a widely used resource for SARS-CoV-2 genetic information. The research, published in *Virus Evolution*, investigated the virus's ability to mutate and infect humans, focusing on deletion repair events in the spike protein, which the virus uses to enter cells. The team discovered that initial findings suggesting frequent deletion repair events—where the virus appears to correct its genome—were largely due to errors in how viral sequences were processed. These errors created the false impression that the virus was repairing mutations more regularly than it actually was. By comparing the database information with raw genome sequencing data, the researchers obtained a more accurate understanding of the virus's genetic changes. Mireia Coscollá Devís, a CSIC researcher and the study leader, noted that the frequency of these deletion repair events was significantly overestimated due to inconsistencies in data processing across different laboratories. The study confirmed that while some repair events are genuine, the majority were artifacts of sequence processing. Less than 60% of the initially reported deletion repair events could be validated, with the corrected frequency estimated to be 5 to 51 times lower than originally indicated. Despite their rarity, these repair events can subtly influence the virus's behavior, potentially affecting how it enters cells or interacts with antibodies produced by vaccination. The study underscores the critical importance of carefully scrutinizing genetic data to avoid drawing inaccurate conclusions about viral evolution. The scientists also highlight the need for a centralized system for sharing pathogen genomic data in Spain to improve monitoring and response to infectious diseases.