feat: added RGI for AMR detection + bug fix

.test/config/config.yml

@@ -33,3 +33,7 @@ panaroo:
 fastani:
   skip: False
   extra: ""
+
+rgi:
+  skip: False
+  extra: "--clean --alignment_tool DIAMOND"

README.md

@@ -35,8 +35,10 @@ _Workflow overview:_
    1. NCBI's Prokaryotic Genome Annotation Pipeline ([PGAP](https://github.com/ncbi/pgap)). Note: needs to be installed manually
    2. [prokka](https://github.com/tseemann/prokka), a fast and light-weight prokaryotic annotation tool
    3. [bakta](https://github.com/oschwengers/bakta), a fast, alignment-free annotation tool. Note: Bakta will automatically download its companion database from zenodo (light: 1.5 GB, full: 40 GB)
-3. Create a QC report for the assemblies using [Quast](https://github.com/ablab/quast)
-4. Create a pangenome analysis (orthologs/homologs) using [Panaroo](https://gthlab.au/panaroo/)
+3. Predict antimicrobial resistance (AMR) genes using [RGI](https://github.com/arpcard/rgi)
+4. Create a QC report for the assemblies using [Quast](https://github.com/ablab/quast)
+5. Create a pangenome analysis (orthologs/homologs) using [Panaroo](https://gthlab.au/panaroo/)
+6. Compute pairwise average nucleotide identity (ANI) between the assemblies using [FastANI](https://github.com/ParBLiSS/FastANI) and plot a phylogenetic tree based on the ANI distances.
 
 ## Installation
 

config/README.md

@@ -7,9 +7,10 @@ A Snakemake workflow for the post-processing of microbial genome assemblies.
    1. NCBI's Prokaryotic Genome Annotation Pipeline ([PGAP](https://github.com/ncbi/pgap)). Note: needs to be installed manually
    2. [prokka](https://github.com/tseemann/prokka), a fast and light-weight prokaryotic annotation tool
    3. [bakta](https://github.com/oschwengers/bakta), a fast, alignment-free annotation tool. Note: Bakta will automatically download its companion database from zenodo (light: 1.5 GB, full: 40 GB)
-3. Create a QC report for the assemblies using [Quast](https://github.com/ablab/quast)
-4. Create a pangenome analysis (orthologs/homologs) using [Panaroo](https://gthlab.au/panaroo/)
-5. Compute pairwise average nucleotide identity (ANI) between the assemblies using [FastANI](https://github.com/ParBLiSS/FastANI) and plot a phylogenetic tree based on the ANI distances.
+3. Predict antimicrobial resistance (AMR) genes using [RGI](https://github.com/arpcard/rgi)
+4. Create a QC report for the assemblies using [Quast](https://github.com/ablab/quast)
+5. Create a pangenome analysis (orthologs/homologs) using [Panaroo](https://gthlab.au/panaroo/)
+6. Compute pairwise average nucleotide identity (ANI) between the assemblies using [FastANI](https://github.com/ParBLiSS/FastANI) and plot a phylogenetic tree based on the ANI distances.
 
 ## Running the workflow
 

config/config.yml

@@ -33,3 +33,7 @@ panaroo:
 fastani:
   skip: False
   extra: ""
+
+rgi:
+  skip: False
+  extra: "--clean --alignment_tool DIAMOND"

config/schemas/config.schema.yml

@@ -133,6 +133,17 @@ properties:
         type: string
         description: Extra command-line arguments for FastANI
         default: ""
+  rgi:
+    type: object
+    properties:
+      skip:
+        type: boolean
+        description: Whether to skip RGI analysis
+        default: false
+      extra:
+        type: string
+        description: Extra command-line arguments for RGI
+        default: "--clean --alignment_tool DIAMOND"
 required:
   - samplesheet
   - tool
@@ -143,3 +154,4 @@ required:
   - quast
   - panaroo
   - fastani
+  - rgi

workflow/rules/common.smk

@@ -50,6 +50,13 @@ def get_panaroo_fasta(wildcards):
 
 def get_final_input(wildcards):
     inputs = []
+    for tool in config["tool"]:
+        inputs += expand(
+            "results/annotation/{tool}/{sample}/{sample}.{ext}",
+            tool=tool,
+            sample=samples.index,
+            ext=["gff", "fna"],
+        )
     inputs += expand(
         "results/qc/quast/report.txt",
     )
@@ -62,6 +69,12 @@ def get_final_input(wildcards):
         inputs += expand(
             "results/qc/fastani/summary.txt",
         )
+    if not config["rgi"]["skip"]:
+        inputs += expand(
+            "results/qc/rgi/{sample}/result.{ext}",
+            sample=samples.index,
+            ext=["txt", "json"],
+        )
     return inputs
 
 

workflow/rules/qc.smk

@@ -124,3 +124,20 @@ rule panaroo:
           {params.extra} \
           > {log} 2>&1
         """
+
+
+rule rgi_detection:
+    input:
+        fasta=rules.get_fasta.output.fasta,
+    output:
+        multiext("results/qc/rgi/{sample}/result", ".txt", ".json"),
+    log:
+        "results/qc/rgi/{sample}/result.log",
+    threads: max(workflow.cores * 0.25, 1)
+    params:
+        input_type="contig",
+        extra=config["rgi"]["extra"],
+    message:
+        """--- Running RGI to detect antibiotic resistance genes ---"""
+    wrapper:
+        "https://raw.githubusercontent.com/MPUSP/mpusp-snakemake-wrappers/refs/heads/main/rgi"