Predicting key viral mutations from within-host deep sequencing data.

Beth-2 detects 98.6% of key SARS-CoV-2 receptor binding domain mutations from within-host deep sequencing, with a median 9-month lead before they become predominant.

• Why it matters: Accurate, timely forecasts of SARS-CoV-2 evolution are needed to support vaccine matching and epidemic response, but high diversity and rapid change make prediction difficult.
• What they did: Beth-2 analyzes early within-host fitness signals and their between-host fitness patterns to infer emerging mutations, then validates retrospectively and tests 26 prospectively predicted mutations.
• The result: Of 26 prospectively predicted mutations, 69% showed significantly enhanced transmissibility, and five showed significant immune escape in plasma from COVID-19 mRNA vaccine recipients sampled 6 months after prediction.

AI summary of the abstract · may contain errors · not medical advice · read the paper.