ingest/nextclade: Deduplicate serotype code

The serotypes still need to be hardcoded because we only have Nextclade
datasets for these specific serotypes. Use these hardcoded serotypes
as wildcard constraints in the Nextclade rules to deduplicate the
shell commands and parallelize the final `split_metadata_by_serotype`
rule.

ingest/workflow/snakemake_rules/nextclade.smk

@@ -1,16 +1,21 @@
+SUPPORTED_NEXTCLADE_SEROTYPES = ['denv1', 'denv2', 'denv3', 'denv4']
+SEROTYPE_CONSTRAINTS = '|'.join(SUPPORTED_NEXTCLADE_SEROTYPES)
+
 rule nextclade_denvX:
     """
     For each type, classify into the appropriate subtype
     """
     input:
-        sequences="results/sequences_denv{x}.fasta",
-        dataset="../nextclade_data/denv{x}",
+        sequences="results/sequences_{serotype}.fasta",
+        dataset="../nextclade_data/{serotype}",
     output:
-        nextclade_denvX="data/nextclade_results/nextclade_denv{x}.tsv",
+        nextclade_denvX="data/nextclade_results/nextclade_{serotype}.tsv",
     threads: 4
     params:
         min_length=config["nextclade"]["min_length"],
         min_seed_cover=config["nextclade"]["min_seed_cover"],
+    wildcard_constraints:
+        serotype=SEROTYPE_CONSTRAINTS
     shell:
         """
         nextclade run \
@@ -28,10 +33,7 @@ rule concat_nextclade_subtype_results:
     Concatenate all the nextclade results for dengue subtype classification
     """
     input:
-        nextclade_denv1="data/nextclade_results/nextclade_denv1.tsv",
-        nextclade_denv2="data/nextclade_results/nextclade_denv2.tsv",
-        nextclade_denv3="data/nextclade_results/nextclade_denv3.tsv",
-        nextclade_denv4="data/nextclade_results/nextclade_denv4.tsv",
+        expand("data/nextclade_results/nextclade_{serotype}.tsv", serotype=SUPPORTED_NEXTCLADE_SEROTYPES),
     output:
         nextclade_subtypes="results/nextclade_subtypes.tsv",
     params:
@@ -43,16 +45,7 @@ rule concat_nextclade_subtype_results:
         | tr ',' '\t' \
         > {output.nextclade_subtypes}
 
-        tsv-select -H -f "seqName,clade" {input.nextclade_denv1} \
-        | awk 'NR>1 {{print}}' \
-        >> {output.nextclade_subtypes}
-        tsv-select -H -f "seqName,clade" {input.nextclade_denv2} \
-        | awk 'NR>1 {{print}}' \
-        >> {output.nextclade_subtypes}
-        tsv-select -H -f "seqName,clade" {input.nextclade_denv3} \
-        | awk 'NR>1 {{print}}' \
-        >> {output.nextclade_subtypes}
-        tsv-select -H -f "seqName,clade" {input.nextclade_denv4} \
+        tsv-select -H -f "seqName,clade" {input} \
         | awk 'NR>1 {{print}}' \
         >> {output.nextclade_subtypes}
         """
@@ -87,14 +80,10 @@ rule split_metadata_by_serotype:
     input:
         metadata="results/metadata_all.tsv",
     output:
-        metadata_denv1="results/metadata_denv1.tsv",
-        metadata_denv2="results/metadata_denv2.tsv",
-        metadata_denv3="results/metadata_denv3.tsv",
-        metadata_denv4="results/metadata_denv4.tsv",
+        serotype_metadata="results/metadata_{serotype}.tsv"
+    wildcard_constraints:
+        serotype=SEROTYPE_CONSTRAINTS
     shell:
         """
-        tsv-filter -H --str-eq ncbi_serotype:denv1 {input.metadata} > {output.metadata_denv1}
-        tsv-filter -H --str-eq ncbi_serotype:denv2 {input.metadata} > {output.metadata_denv2}
-        tsv-filter -H --str-eq ncbi_serotype:denv3 {input.metadata} > {output.metadata_denv3}
-        tsv-filter -H --str-eq ncbi_serotype:denv4 {input.metadata} > {output.metadata_denv4}
+        tsv-filter -H --str-eq ncbi_serotype:{wildcards.serotype} {input.metadata} > {output.serotype_metadata}
         """