Merge pull request #332 from pangenome/sync_partition_before_pggb

update `partition-before-pggb`
pangenome · Sep 28, 2023 · 3c17da0 · 3c17da0
2 parents 564586e + 1f53de2
commit 3c17da0
Show file tree

Hide file tree

Showing 2 changed files with 31 additions and 10 deletions.
diff --git a/partition-before-pggb b/partition-before-pggb
@@ -87,22 +87,45 @@ no_merge_segments=false
 block_ratio_min=0
 reduce_redundancy=true
 
+# function that parses metric suffix numerical arguments
+parse_numeric() {
+  local input=$1
+  local value=${input%[KkMmGgTt]*}
+  local suffix=${input#$value}
+  if [[ -z $suffix ]]; then
+    if [[ $value =~ ^[0-9]+(\.[0-9]+)?$ ]]; then
+      echo $value
+      return 0
+    else
+      echo "[pggb] ERROR: Invalid input" >&2
+      return 1
+    fi
+  fi
+  case $suffix in
+    K|k) value=$(echo "$value * 1000" | bc) ;;
+    M|m) value=$(echo "$value * 1000000" | bc) ;;
+    G|g) value=$(echo "$value * 1000000000" | bc) ;;
+    T|t) value=$(echo "$value * 1000000000000" | bc) ;;
+    *) echo "[pggb] ERROR: Invalid suffix or unsupported suffix. Supported metric suffixes are k, K, m, M, g, G, t, T." >&2; return 1 ;;
+  esac
+  printf "%.0f" $value
+}
 
 if [ $# -eq 0 ]; then
     show_help=true
 fi
 
 # read the options
 cmd=$0" "$@
-TEMP=`getopt -o i:o:D:a:p:n:s:l:K:F:k:x:f:B:XH:j:P:O:Me:t:T:vhASY:G:Q:d:I:R:NbrmZzV: --long input-fasta:,output-dir:,temp-dir:,input-paf:,map-pct-id:,n-mappings:,segment-length:,block-length-min:,mash-kmer:,mash-kmer-thres:,min-match-length:,sparse-map:,sparse-factor:,transclose-batch:,skip-normalization,n-haps:,path-jump-max:,subpath-min:,edge-jump-max:,threads:,poa-threads:,skip-viz,do-layout,help,no-merge-segments,do-stats,exclude-delim:,poa-length-target:,poa-params:,poa-padding:,run-abpoa,global-poa,write-maf,consensus-spec:,consensus-prefix:,pad-max-depth:,block-id-min:,block-ratio-min:,no-splits,resume,keep-temp-files,multiqc,compress,vcf-spec:,version -n 'pggb' -- "$@"`
+TEMP=`getopt -o i:o:D:a:p:n:s:l:K:F:k:x:f:B:XH:j:P:O:Me:t:T:vhASY:G:Q:d:I:R:NbrmZzV: --long input-fasta:,output-dir:,temp-dir:,input-paf:,map-pct-id:,n-mappings:,segment-length:,block-length-min:,mash-kmer:,mash-kmer-thres:,min-match-length:,sparse-map:,sparse-factor:,transclose-batch:,skip-normalization,n-haps:,path-jump-max:,subpath-min:,edge-jump-max:,threads:,poa-threads:,skip-viz,do-layout,help,no-merge-segments,stats,exclude-delim:,poa-length-target:,poa-params:,poa-padding:,run-abpoa,global-poa,write-maf,consensus-spec:,consensus-prefix:,pad-max-depth:,block-id-min:,block-ratio-min:,no-splits,resume,keep-temp-files,multiqc,compress,vcf-spec:,version -n 'pggb' -- "$@"`
 eval set -- "$TEMP"
 
 # extract options and their arguments into variables.
 while true ; do
     case "$1" in
         -i|--input-fasta) input_fasta=$2 ; shift 2 ;;
-        -s|--segment-length) segment_length=$2 ; shift 2 ;;
-        -l|--block-length) block_length=$2 ; shift 2 ;;
+        -s|--segment-length) segment_length=$(parse_numeric $2) ; shift 2 ;;
+        -l|--block-length) block_length=$(parse_numeric $2) ; shift 2 ;;
         -p|--map-pct-id) map_pct_id=$2 ; shift 2 ;;
         -n|--n-haplotypes) n_mappings=$2 ; shift 2 ;;
         -N|--no-splits) no_splits=true ; shift ;;
@@ -112,7 +135,7 @@ while true ; do
         -Y|--exclude-delim) exclude_delim=$2 ; shift 2 ;;
         -k|--min-match-length) min_match_length=$2 ; shift 2 ;;
         -f|--sparse-factor) sparse_factor=$2 ; shift 2 ;;
-        -B|--transclose-batch) transclose_batch=$2 ; shift 2 ;;
+        -B|--transclose-batch) transclose_batch=$(parse_numeric $2) ; shift 2 ;;
         -X|--skip-normalization) skip_normalization=true ; shift ;;
         -H|--n-haplotypes-smooth) n_haps=$2 ; shift 2 ;;
         -j|--path-jump-max) max_path_jump=$2 ; shift 2 ;;
@@ -127,7 +150,7 @@ while true ; do
         #-C|--consensus-spec) consensus_spec=$2 ; shift 2 ;;
         -Q|--consensus-prefix) consensus_prefix=$2 ; shift 2 ;;
         -v|--skip-viz) do_viz=false ; do_layout=false; shift ;;
-        -S|--do-stats) do_stats=true ; shift ;;
+        -S|--stats) do_stats=true ; shift ;;
         -V|--vcf-spec) vcf_spec=$2 ; shift 2 ;;
         -m|--multiqc) multiqc=true ; shift ;;
         -o|--output-dir) output_dir=$2 ; shift 2 ;;
@@ -148,7 +171,7 @@ done
 if [ $show_version == true ]; then
     SCRIPT_DIR=$( cd -- "$(dirname -- "$(readlink -f "${BASH_SOURCE[0]}" )" )" &> /dev/null && pwd )
     cd "$SCRIPT_DIR"
-    GIT_VERSION=$(git describe --always --tags)
+    GIT_VERSION=$(git describe --always --tags --long)
     echo "pggb $GIT_VERSION"
     cd - &> /dev/null
     exit
@@ -356,7 +379,6 @@ if [ ! -e "$temp_dir" ]; then
   mkdir "$temp_dir"
   temp_dir_was_created=true
 fi
-
 prefix_mappings_paf="$temp_dir"/$(basename "$prefix_paf")
 prefix_seqwish="$temp_dir"/$(basename "$prefix_seqwish")
 prefix_smoothed="$temp_dir"/$(basename "$prefix_smoothed")

diff --git a/pggb b/pggb
@@ -454,9 +454,6 @@ reporting:
   multiqc:            $multiqc
 EOT
 
-
-echo -e "\nRunning pggb\n" >> "$log_file"
-
 # Check Pangenome Sequence Naming (PanSN)
 if [ ! -f "${input_fasta}.fai" ]; then
     echo "[pggb] ERROR: Index for $input_fasta does not exist. Please create it using 'samtools faidx $input_fasta'."
@@ -470,6 +467,8 @@ cut -f 1 "${input_fasta}.fai" | while read -r line; do
   fi
 done
 
+echo -e "\nRunning pggb\n" >> "$log_file"
+
 if [[ "$input_paf" == false ]]; then
   if [[ ! -s "$prefix_paf".alignments.$mapper.paf || $resume == false ]]; then