baa98a4c7bb930fd0bf6365f92436eb900ade8e3
angie
  Sat Mar 12 20:31:52 2022 -0800
Prevent duplicate lines in ncbi_dataset.plusBioSample.tsv.

diff --git src/hg/utils/otto/sarscov2phylo/getNcbi.sh src/hg/utils/otto/sarscov2phylo/getNcbi.sh
index 542f85f..7352205 100755
--- src/hg/utils/otto/sarscov2phylo/getNcbi.sh
+++ src/hg/utils/otto/sarscov2phylo/getNcbi.sh
@@ -40,34 +40,34 @@
         retryDelay=$(($retryDelay * 2))
     fi
 done
 if [[ ! -f ncbi_dataset.zip ]]; then
     echo "datasets command failed $maxAttempts times; quitting."
     exit 1
 fi
 rm -rf ncbi_dataset
 unzip -o ncbi_dataset.zip
 # Creates ./ncbi_dataset/
 
 # This makes something just like ncbi.datasets.tsv from the /table/ API query:
 jq -c -r '[.accession, .biosample, .isolate.collectionDate, .location.geographicLocation, .host.sciName, .isolate.name, .completeness, (.length|tostring)] | join("\t")' \
     ncbi_dataset/data/data_report.jsonl \
 | sed -e 's/COMPLETE/complete/; s/PARTIAL/partial/;' \
-| sort \
+| sort -u \
     > ncbi_dataset.tsv
 
-time $scriptDir/bioSampleJsonToTab.py ncbi_dataset/data/biosample.jsonl > gb.bioSample.tab
+time $scriptDir/bioSampleJsonToTab.py ncbi_dataset/data/biosample.jsonl | uniq > gb.bioSample.tab
 
 # Use BioSample metadata to fill in missing pieces of GenBank metadata and report conflicting
 # sample collection dates:
 $scriptDir/gbMetadataAddBioSample.pl gb.bioSample.tab ncbi_dataset.tsv \
     > ncbi_dataset.plusBioSample.tsv 2>gbMetadataAddBioSample.log
 
 # Make a file for joining collection date with ID:
 tawk '$3 != "" {print $1, $3;}' ncbi_dataset.plusBioSample.tsv \
 | sort > gbToDate
 
 # Replace FASTA headers with reconstructed names from enhanced metadata.
 time cleanGenbank < ncbi_dataset/data/genomic.fna \
 | $scriptDir/fixNcbiFastaNames.pl ncbi_dataset.plusBioSample.tsv \
 | xz -T 20 \
     > genbank.fa.xz