custom reference database taxonomic assignment rule

rules/custom_taxonomic_assignment.smk

+__author__ = "Pierre-Edouard Guerin"
+__license__ = "MIT"
+
+
+## Assign each sequence to a taxon
+rule custom_taxonomic_assignment:
+    input:
+        'results/14_dereplicate_runs/{projmarkrun}.uniq.fasta'
+    output:
+        'results/15_taxonomic_assignment/{projmarkrun}.tag.u.fasta'
+    threads:
+        workflow.cores * 0.5
+    conda:
+        '../envs/obitools_envs.yaml'
+    singularity:
+        config["singularity"]["obitools"]
+    params:
+        drun= lambda wildcards: dfpmr[dfpmr.projmarkrun == wildcards.projmarkrun].to_dict('records')[0],        
+    log:
+        'logs/15_taxonomic_assignment/{projmarkrun}.log'
+    shell:
+        '''
+        mkdir -p results/15_taxonomic_assignment/{params.drun[projet]}/{params.drun[marker]};      
+        SUB="_custom"
+        STR={input}
+        if [[ "$STR" == *"$SUB"* ]]; then
+         ecotag -t {params.drun[bdr]} -R {params.drun[fasta]} {input} > {output} 2> {log}        
+        else         
+         ecotag -d {params.drun[bdr]} -R {params.drun[fasta]} {input} > {output} 2> {log}
+        fi
+        '''
\ No newline at end of file