Estimating rates of SARS-CoV-2 lineage spread using graph spectra from genetic sequence data.

Graph-Laplacian features of genetic diversity and graph connectivity from early SARS-CoV-2 sequence data (US and UK, 2021–mid-2022) reliably estimate growth rates of newly emerged lineages.

• Why it matters: Rapidly growing lineages can be hard to detect within massive, streaming genomic datasets, limiting real-time epidemic inference and monitoring.
• What they did: The method encodes distances between small sequence subsets as graph Laplacians, then uses genetic diversity and graph connectivity to infer growth rates from early US/UK data (2021–mid-2022).
• The result: In later US/UK data (mid-2022 to end-2024), the model accurately inferred lineage growth rates and, in 2023 with thousands of sequences per day, correctly identified the most rapidly expanding lineages while still at low frequencies.

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