feat: include previous clustering

RIVM-bioinformatics · Jun 28, 2024 · f2767d0 · f2767d0
1 parent 00f759e
commit f2767d0
Show file tree

Hide file tree

Showing 4 changed files with 21 additions and 8 deletions.
diff --git a/Snakefile b/Snakefile
@@ -11,6 +11,10 @@ for param in ["threads", "mem_gb"]:
 
 OUT = config["output_dir"]
 
+# find collection using collfinder
+# iget collection and save as "previous_clustering" in working dir
+PREVIOUS_CLUSTERING = "previous_clustering"
+
 # Configure pipeline outputs
 expected_outputs = []
 

diff --git a/workflow/rules/clustering.smk b/workflow/rules/clustering.smk
@@ -9,7 +9,7 @@ rule clustering:
     message:
         "Clustering {input.distances} with threshold {params.threshold}"
     resources:
-        mem_gb=config["mem_gb"]["compression"],
+        mem_gb=config["mem_gb"]["clustering"],
     conda:
         "../envs/clustering.yaml"
     container:

diff --git a/workflow/rules/combine_snp_profiles.smk b/workflow/rules/combine_snp_profiles.smk
@@ -3,7 +3,7 @@
 
 rule decompress_snp_profiles:
     input:
-        INPUT + "/aln.fa.gz",
+        PREVIOUS_CLUSTERING + "/aln.fa.gz",
     output:
         temp(OUT + "/old_aln.fa"),
     log:
@@ -15,7 +15,7 @@ rule decompress_snp_profiles:
     conda:
         "../envs/scripts.yaml"
     container:
-        ""
+        "docker://ghcr.io/boasvdp/juno_clustering_scripts:0.1"
     params:
         script="workflow/scripts/script.py",
     threads: config["threads"]["compression"]
@@ -40,13 +40,15 @@ rule add_snp_profiles:
     message:
         "Adding SNP profiles to {input.previous_aln}."
     resources:
-        mem_gb=config["mem_gb"]["add_snp_profiles"],
+        mem_gb=config["mem_gb"]["compression"],
     conda:
         "../envs/scripts.yaml"
     container:
-        ""
+        "docker://ghcr.io/boasvdp/juno_clustering_scripts:0.1"
+    threads: config["threads"]["compression"]
     shell:
         """
+# TODO: find better way of combining samples, e.g. make sure no duplicate names
 cat {input.previous_aln} {input.assemblies} > {output}
         """
 
@@ -65,7 +67,7 @@ rule compress_snp_profiles:
     conda:
         "../envs/scripts.yaml"
     container:
-        ""
+        "docker://ghcr.io/boasvdp/juno_clustering_scripts:0.1"
     threads: config["threads"]["compression"]
     shell:
         """

diff --git a/workflow/rules/distance_calculation.smk b/workflow/rules/distance_calculation.smk
@@ -6,16 +6,20 @@ rule distance_calculation_snp:
     output:
         OUT + "/distances.tsv",
     conda:
-        ""
+        "../envs/distance_calculation.yaml"
     container:
         "docker://ghcr.io/boasvdp/distle:0.1.0"
     params:
         max_distance=config["max_distance"],
         output_mode="full",
+    resources:
+        mem_gb=config["mem_gb"]["distance_calculation"],
     log:
         OUT + "/log/distance_calculation_snp.log",
+    threads: config["threads"]["distance_calculation"]
     shell:
         """
+# TODO: check if the distances have already been calculated and only calculate the missing ones, this could be added to distle
 distle \
 --verbose \
 --input-format fasta \
@@ -31,14 +35,17 @@ rule distance_calculation_cgmlst:
     output:
         OUT + "/distances.tsv",
     conda:
-        ""
+        "../envs/distance_calculation.yaml"
     container:
         "docker://ghcr.io/boasvdp/distle:0.1.0"
     params:
         max_distance=config["max_distance"],
         output_mode="full",
+    resources:
+        mem_gb=config["mem_gb"]["distance_calculation"],
     log:
         OUT + "/log/distance_calculation_cgmlst.log",
+    threads: config["threads"]["distance_calculation"]
     shell:
         """
 distle \