c8d7abe94b3244702501b71b62d9e1ea6333228b angie Sat Jan 23 14:24:20 2021 -0800 Grab COG-UK daily tree as Russ suggests; might want to try using it instead of GISAID tree. diff --git src/hg/utils/otto/sarscov2phylo/getCogUk.sh src/hg/utils/otto/sarscov2phylo/getCogUk.sh index b2affda..f17146f 100755 --- src/hg/utils/otto/sarscov2phylo/getCogUk.sh +++ src/hg/utils/otto/sarscov2phylo/getCogUk.sh @@ -5,21 +5,22 @@ # Download latest COG-UK fasta and metadata; update $ottoDir/cogUk.latest link. scriptDir=$(dirname "${BASH_SOURCE[0]}") source $scriptDir/util.sh today=$(date +%F) ottoDir=/hive/data/outside/otto/sarscov2phylo cogUrlBase=https://cog-uk.s3.climb.ac.uk/phylogenetics/latest mkdir -p $ottoDir/cogUk.$today cd $ottoDir/cogUk.$today curl -S -s $cogUrlBase/cog_all.fasta | xz -T 50 > cog_all.fasta.xz curl -S -s $cogUrlBase/cog_metadata.csv > cog_metadata.csv +curl -S -s $cogUrlBase/cog_global_tree.newick > cog_global_tree.newick tail -n +2 cog_metadata.csv \ | awk -F, '{print $1 "\t" $4;}' | sort > cogUkToDate rm -f $ottoDir/cogUk.latest ln -s cogUk.$today $ottoDir/cogUk.latest