A Global Comparison of COVID-19 Variant Waves and Relationships with Clinical and...
summary
Global analysis of 11.5 million sequences from GISAID using PANGO lineages fits a flexible multinomial model with cubic splines (better than linear fits) to summarize variant transitions by three summaries: speed, timing, magnitude. Analysis covers 215 locations out of 590, quantifies Delta as an inflection point and Omicron’s different behavior, and uses random forests to show top drivers are vaccination, number of co‑circulating variants, and cumulative cases per million. The pipeline characterizes transitions robustly but is not yet integrated into operational case forecasts.
AI-written summary · about
chapters (8)
- 0:00Introduction and background
- 2:06Project motivation and goals
- 7:33Data sources and heterogeneity
- 11:22Methodology for variant analysis
- 16:05Key metrics: speed and timing
- 23:10Impact of external factors
- 30:44Clustering and takeaways
- 34:29Q&A and future directions
related talks
34mAn integrated genomic surveillance platform reveals multiple introductions of SARS-CoV-2
33mApplication of Multi-class Supervised Algorithms to Identify COVID-19 Viral Strains
43mProbabilistic Models for Pathogen Evolution Within and Between Hosts
31mBayesian Risk Analysis For Large Multi-Modal Data
27mBeyond a Machine Learning Model
27mA Statistical, Reference-Free Algorithm Subsumes Myriad Problems in Genome Science
28mMachine Learning vs The Flu
34mLongitudinal Phenotypes and Disease Trajectories at Population Scale