3760b502d693a5e3e1540f5eb190c87a6a0e965f
angie
  Tue Feb 22 11:04:44 2022 -0800
Strip quotes from nextclade tsv output, speed up compression by using more threads & speed up sed by usin conditional.

diff --git src/hg/utils/otto/sarscov2phylo/gisaidFromChunks.sh src/hg/utils/otto/sarscov2phylo/gisaidFromChunks.sh
index 235aa79..418f74a 100755
--- src/hg/utils/otto/sarscov2phylo/gisaidFromChunks.sh
+++ src/hg/utils/otto/sarscov2phylo/gisaidFromChunks.sh
@@ -15,44 +15,45 @@
 today=$(date +%F)
 
 # Run pangolin and nextclade on any chunks that need it
 cd /hive/users/angie/gisaid/chunks
 make nextclade.tsv
 make -j10
 
 cd /hive/users/angie/gisaid
 # Glom all the chunks together.
 # Remove initial "hCoV-19/" and remove spaces a la nextmeta (e.g. "Hong Kong" -> "HongKong").
 # Strip single quotes (e.g. "Cote d'Ivoire" --> "CotedIvoire").
 # Also remove a stray comma in a name that caused Newick parsing error ("Hungary/US-32533w,/2020").
 # Keep the strain|epiId|date "full names".
 time xzcat chunks/gisaid_epi_isl_*.fa.xz \
 | sed -re 's@^>hCo[Vv]-19/+@>@;  s/[ '"'"',()]//g;  s/\r$//;' \
-| xz -T 8 \
+| xz -T 20 \
     > gisaid_fullNames_$today.fa.xz
 
 # Make tmp files with a fullName key and various columns that we'll join together.
 fastaNames gisaid_fullNames_$today.fa.xz \
 | awk -F\| -vOFS="\t" '{print $0, $1, $2, $3;}' \
 | sort -u \
     > tmp.first3
 # Sequence length
 faSize -detailed  <(xzcat gisaid_fullNames_$today.fa.xz) | sort -u > tmp.lengths
 # Lineage & clade assignments
 sort -u chunks/pangolin.tsv \
     > tmp.lineage
-sort -u chunks/nextclade.tsv \
+sed -re 's/"//g;' chunks/nextclade.tsv \
+| sort -u \
     > tmp.clade
 # Countries -- go back to unstripped sequence names:
 xzcat chunks/gisaid_epi_isl_*.fa.xz \
 | grep ^\> \
 | sed -re 's@^>hCo[Vv]-19/+@@;' \
 | $scriptDir/gisaidNameToCountry.pl \
 | sort -u \
     > tmp.country
 
 # Join locally computed fields and sort by EPI ID for joining with latest real nextmeta
 join -t$'\t' -a 1 tmp.first3 tmp.lengths \
 | join -t$'\t' -a 1 -o 1.1,1.2,1.3,1.4,1.5,2.2 - tmp.clade \
 | join -t$'\t' -a 1 -o 1.1,1.2,1.3,1.4,1.5,1.6,2.2 - tmp.lineage \
 | join -t$'\t' -a 1 -o 1.1,1.2,1.3,1.4,1.5,1.6,1.7,2.2 - tmp.country \
 | tawk '{print $3, $2, $4, $5, $6, $7, $8;}' \
@@ -70,25 +71,25 @@
 #19      pangolin_lineage
 # Fill in other columns from nextmeta when possible (join on EPI ID since names change over time)
 set +o pipefail
 zcat $lastRealNextmeta | head -1 \
     > metadata_batch_$today.tsv
 set -o pipefail
 zcat $lastRealNextmeta \
 | tail -n+2 \
 | sort -k 3,3 \
 | join -t$'\t' -a 1 -2 3 \
     -o 1.2,2.2,1.1,2.4,1.3,2.6,1.7,2.8,2.9,2.10,2.11,2.12,2.13,1.4,2.15,2.16,2.17,1.5,1.6,2.20,2.21,2.22,2.23,2.24,2.25,2.26,2.27 \
     tmp.epiToLocalMeta - \
 | sort \
     >> metadata_batch_$today.tsv
 wc -l metadata_batch_$today.tsv
-gzip -f metadata_batch_$today.tsv
+pigz -p 8 -f metadata_batch_$today.tsv
 
 # Make fasta with strain-name headers a la nextfasta.
 xzcat gisaid_fullNames_$today.fa.xz \
-| sed -re 's/\|.*//' \
-| xz -T 8 \
+| sed -re '/^>/ s/\|.*//' \
+| xz -T 20 \
     > sequences_batch_$today.fa.xz
 
 # Clean up
 rm tmp.*