71a5bf4d42948792ba96729c01bec373eed86918
angie
  Sun Feb 4 09:16:38 2024 -0800
Use nextclade v3 output column indices.

diff --git src/hg/utils/otto/sarscov2phylo/getCogUk.sh src/hg/utils/otto/sarscov2phylo/getCogUk.sh
index be1c1af..85ee6c0 100755
--- src/hg/utils/otto/sarscov2phylo/getCogUk.sh
+++ src/hg/utils/otto/sarscov2phylo/getCogUk.sh
@@ -42,31 +42,31 @@
 
 curlRetry $cogUrlBase/cog_all.fasta.gz
 curlRetry $cogUrlBase/cog_metadata.csv.gz
 curlRetry $cogUrlBase/cog_global_tree.newick
 
 zcat cog_all.fasta.gz | xz -T 20 > cog_all.fasta.xz
 rm cog_all.fasta.gz
 
 zcat cog_metadata.csv.gz \
 | tail -n +2 \
 | awk -F, '{print $1 "\t" $5;}' | sort > cogUkToDate
 
 # Reuse nextclade assignments for older sequences; compute nextclade assignments for new seqs.
 zcat $ottoDir/cogUk.latest/nextclade.full.tsv.gz > nextclade.full.tsv
 cp $ottoDir/cogUk.latest/nextalign.fa.xz .
-comm -13 <(cut -f 1 nextclade.full.tsv | sort) <(fastaNames cog_all.fasta.xz | sort) \
+comm -13 <(cut -f 2 nextclade.full.tsv | sort) <(fastaNames cog_all.fasta.xz | sort) \
     > seqsForNextclade
 if [ -s seqsForNextclade ]; then
     nDataDir=~angie/github/nextclade/data/sars-cov-2
     outTsv=$(mktemp)
     outFa=$(mktemp)
     faSomeRecords <(xzcat cog_all.fasta.xz) seqsForNextclade stdout \
     | nextclade run -j 30 \
         --input-dataset $nDataDir \
         --output-fasta $outFa \
         --output-tsv $outTsv >& nextclade.log
     tail -n+2 $outTsv | sed -re 's/"//g;' >> nextclade.full.tsv
     xz -T 20 < $outFa >> nextalign.fa.xz
     rm -f $outTsv $outFa
 fi
 pigz -f -p 8 nextclade.full.tsv