Merge pull request #372 from pangenome/naming

add option for filenames with param values
pangenome · Feb 3, 2024 · 6ffe7f9 · 6ffe7f9
2 parents 34de9a0 + 14c7c45
commit 6ffe7f9
Show file tree

Hide file tree

Showing 2 changed files with 40 additions and 10 deletions.
diff --git a/partition-before-pggb b/partition-before-pggb
@@ -27,7 +27,7 @@ exclude_delim="#"
 # seqwish's default values
 MIN_MATCH_LENGTH=23
 SPARSE_FACTOR=0
-TRANSCLOSE_BATCH=10000000
+TRANSCLOSE_BATCH=10M
 
 # seqwish's parameters
 min_match_length=$MIN_MATCH_LENGTH
@@ -80,6 +80,7 @@ threads=$THREADS
 poa_threads=0
 keep_intermediate_files=false
 compress=false
+names_with_params=false
 show_version=false
 show_help=false
 
@@ -118,7 +119,7 @@ fi
 
 # read the options
 cmd=$0" "$@
-TEMP=`getopt -o i:o:D:a:p:c:g:s:l:K:F:k:x:f:B:Xn: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:,hg-filter-ani-diff:,segment-length:,block-length-min:,mash-kmer:,mash-kmer-thres:,min-match-length:,sparse-map:,sparse-factor:,transclose-batch:,skip-normalization,n-haplotypes:,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' -- "$@"`
+TEMP=`getopt -o i:o:D:a:p:c:g:s:l:K:F:k:x:f:B:Xn:j:P:O:Me:t:T:vhASY:G:Q:d:I:R:NbrmZzV: --long input-fasta:,output-dir:,temp-dir:,input-paf:,map-pct-id:,n-mappings:,hg-filter-ani-diff:,segment-length:,block-length-min:,mash-kmer:,mash-kmer-thres:,min-match-length:,sparse-map:,sparse-factor:,transclose-batch:,skip-normalization,n-haplotypes:,path-jump-max:,subpath-min:,edge-jump-max:,threads:,poa-threads:,skip-viz,do-layout,help,no-merge-segments,stats,exclude-delim:,poa-length-target:,poa-params:,poa-padding:,run-abpoa,global-poa,write-maf,consensus-spec:,consensus-prefix:,pad-max-depth:,block-id-min:,block-ratio-min:,no-splits,resume,keep-temp-files,multiqc,compress,names-with-params,vcf-spec:,version -n 'pggb' -- "$@"`
 eval set -- "$TEMP"
 
 # extract options and their arguments into variables.
@@ -163,6 +164,7 @@ while true ; do
         -T|--poa-threads) poa_threads=$2 ; shift 2 ;;
         -A|--keep-temp-files) keep_intermediate_files=true ; shift ;;
         -Z|--compress) compress=true ; shift ;;
+        --names-with-params) names_with_params=true ; shift ;;
         --version) show_version=true ; shift ;;
         -h|--help) show_help=true ; shift ;;
         --) shift ; break ;;
@@ -249,6 +251,7 @@ if [ "$show_help" == true ]; then
     echo "    -A, --keep-temp-files       keep intermediate graphs"
     echo "    -Z, --compress              compress alignment (.paf), graph (.gfa, .og), and MSA (.maf) outputs with pigz,"
     echo "                                and variant (.vcf) outputs with bgzip"
+    echo "    --names-with-params         put parameter values in filenames, instead of hashes"
     echo "    --version                   display the version of pggb"
     echo "    -h, --help                  this text"
     echo
@@ -347,14 +350,22 @@ else
 fi
 
 if [[ "$input_paf" == false ]]; then
-    prefix_paf="$input_fasta".$(echo "$paf_spec" | sha256sum | head -c 7)
+    if [[ "$names_with_params" == true ]]; then
+        prefix_paf="$input_fasta"."$paf_spec"
+    else
+        prefix_paf="$input_fasta".$(echo "$paf_spec" | sha256sum | head -c 7)
+    fi
 else
     prefix_paf="$input_paf"
 fi
 
 
 # Graph induction
-prefix_seqwish="$prefix_paf".$(echo k$min_match_length-f$sparse_factor-B$transclose_batch | sha256sum | head -c 7)
+if [[ "$names_with_params" == true ]]; then
+    prefix_seqwish="$prefix_paf".k$min_match_length-f$sparse_factor-B$transclose_batch
+else
+    prefix_seqwish="$prefix_paf".$(echo k$min_match_length-f$sparse_factor-B$transclose_batch | sha256sum | head -c 7)
+fi
 
 # poa param suggestions from minimap2
 # - asm5, --poa-params 1,19,39,3,81,1, ~0.1 divergence
@@ -379,7 +390,11 @@ else
 fi
 
 block_id_min=$(echo "scale=4; $map_pct_id / 100.0" | bc)
-prefix_smoothed="$prefix_seqwish".$(echo h$n_haps-G$target_poa_length-j$max_path_jump-e$max_edge_jump-d$pad_max_depth-I$block_id_min-R$block_ratio_min-p$poa_params-O$poa_padding | sha256sum | head -c 7).smooth
+if [[ "$names_with_params" == true ]]; then
+    prefix_smoothed="$prefix_seqwish".h$n_haps-G$target_poa_length-j$max_path_jump-e$max_edge_jump-d$pad_max_depth-I$block_id_min-R$block_ratio_min-p$poa_params-O$poa_padding.smooth
+else
+    prefix_smoothed="$prefix_seqwish".$(echo h$n_haps-G$target_poa_length-j$max_path_jump-e$max_edge_jump-d$pad_max_depth-I$block_id_min-R$block_ratio_min-p$poa_params-O$poa_padding | sha256sum | head -c 7).smooth
+fi
 prefix_smoothed_output="$prefix_smoothed"
 
 

diff --git a/pggb b/pggb
@@ -27,7 +27,7 @@ exclude_delim="#"
 # seqwish's default values
 MIN_MATCH_LENGTH=23
 SPARSE_FACTOR=0
-TRANSCLOSE_BATCH=10000000
+TRANSCLOSE_BATCH=10M
 
 # seqwish's parameters
 min_match_length=$MIN_MATCH_LENGTH
@@ -80,6 +80,7 @@ threads=$THREADS
 poa_threads=0
 keep_intermediate_files=false
 compress=false
+names_with_params=false
 show_version=false
 show_help=false
 
@@ -118,7 +119,7 @@ fi
 
 # read the options
 cmd=$0" "$@
-TEMP=`getopt -o i:o:D:a:p:c:g:s:l:K:F:k:x:f:B:Xn: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:,hg-filter-ani-diff:,segment-length:,block-length-min:,mash-kmer:,mash-kmer-thres:,min-match-length:,sparse-map:,sparse-factor:,transclose-batch:,skip-normalization,n-haplotypes:,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' -- "$@"`
+TEMP=`getopt -o i:o:D:a:p:c:g:s:l:K:F:k:x:f:B:Xn:j:P:O:Me:t:T:vhASY:G:Q:d:I:R:NbrmZzV: --long input-fasta:,output-dir:,temp-dir:,input-paf:,map-pct-id:,n-mappings:,hg-filter-ani-diff:,segment-length:,block-length-min:,mash-kmer:,mash-kmer-thres:,min-match-length:,sparse-map:,sparse-factor:,transclose-batch:,skip-normalization,n-haplotypes:,path-jump-max:,subpath-min:,edge-jump-max:,threads:,poa-threads:,skip-viz,do-layout,help,no-merge-segments,stats,exclude-delim:,poa-length-target:,poa-params:,poa-padding:,run-abpoa,global-poa,write-maf,consensus-spec:,consensus-prefix:,pad-max-depth:,block-id-min:,block-ratio-min:,no-splits,resume,keep-temp-files,multiqc,compress,names-with-params,vcf-spec:,version -n 'pggb' -- "$@"`
 eval set -- "$TEMP"
 
 # extract options and their arguments into variables.
@@ -163,6 +164,7 @@ while true ; do
         -T|--poa-threads) poa_threads=$2 ; shift 2 ;;
         -A|--keep-temp-files) keep_intermediate_files=true ; shift ;;
         -Z|--compress) compress=true ; shift ;;
+        --names-with-params) names_with_params=true ; shift ;;
         --version) show_version=true ; shift ;;
         -h|--help) show_help=true ; shift ;;
         --) shift ; break ;;
@@ -249,6 +251,7 @@ if [ "$show_help" == true ]; then
     echo "    -A, --keep-temp-files       keep intermediate graphs"
     echo "    -Z, --compress              compress alignment (.paf), graph (.gfa, .og), and MSA (.maf) outputs with pigz,"
     echo "                                and variant (.vcf) outputs with bgzip"
+    echo "    --names-with-params         put parameter values in filenames, instead of hashes"
     echo "    --version                   display the version of pggb"
     echo "    -h, --help                  this text"
     echo
@@ -347,14 +350,22 @@ else
 fi
 
 if [[ "$input_paf" == false ]]; then
-    prefix_paf="$input_fasta".$(echo "$paf_spec" | sha256sum | head -c 7)
+    if [[ "$names_with_params" == true ]]; then
+        prefix_paf="$input_fasta"."$paf_spec"
+    else
+        prefix_paf="$input_fasta".$(echo "$paf_spec" | sha256sum | head -c 7)
+    fi
 else
     prefix_paf="$input_paf"
 fi
 
 
 # Graph induction
-prefix_seqwish="$prefix_paf".$(echo k$min_match_length-f$sparse_factor-B$transclose_batch | sha256sum | head -c 7)
+if [[ "$names_with_params" == true ]]; then
+    prefix_seqwish="$prefix_paf".k$min_match_length-f$sparse_factor-B$transclose_batch
+else
+    prefix_seqwish="$prefix_paf".$(echo k$min_match_length-f$sparse_factor-B$transclose_batch | sha256sum | head -c 7)
+fi
 
 # poa param suggestions from minimap2
 # - asm5, --poa-params 1,19,39,3,81,1, ~0.1 divergence
@@ -379,7 +390,11 @@ else
 fi
 
 block_id_min=$(echo "scale=4; $map_pct_id / 100.0" | bc)
-prefix_smoothed="$prefix_seqwish".$(echo h$n_haps-G$target_poa_length-j$max_path_jump-e$max_edge_jump-d$pad_max_depth-I$block_id_min-R$block_ratio_min-p$poa_params-O$poa_padding | sha256sum | head -c 7).smooth
+if [[ "$names_with_params" == true ]]; then
+    prefix_smoothed="$prefix_seqwish".h$n_haps-G$target_poa_length-j$max_path_jump-e$max_edge_jump-d$pad_max_depth-I$block_id_min-R$block_ratio_min-p$poa_params-O$poa_padding.smooth
+else
+    prefix_smoothed="$prefix_seqwish".$(echo h$n_haps-G$target_poa_length-j$max_path_jump-e$max_edge_jump-d$pad_max_depth-I$block_id_min-R$block_ratio_min-p$poa_params-O$poa_padding | sha256sum | head -c 7).smooth
+fi
 prefix_smoothed_output="$prefix_smoothed"