Merge pull request #10 from cbg-ethz/master

Merging new changes from cbg-ethz/v-pipe

workflow/envs/bcftools.yaml

@@ -2,5 +2,5 @@ channels:
   - conda-forge
   - bioconda
 dependencies:
-  - bcftools = 1.13
-  - cyvcf2 = 0.30.11
+  - bcftools = 1.20
+  - cyvcf2 = 0.31.0

workflow/envs/lollipop.yaml

@@ -2,4 +2,4 @@ channels:
   - conda-forge
   - bioconda
 dependencies:
-  - lollipop =0.3.0
+  - lollipop =0.4.1

workflow/rules/common.smk

@@ -685,11 +685,20 @@ for srec in sample_list:
         # in adition to standard VCF files, ShoRAH2 also produces CSV tables
         if config.general["snv_caller"] == "shorah":
             results.append(os.path.join(sdir, "variants/SNVs/snvs.csv"))
+        elif config.general["snv_caller"] == "viloca":
+            results.append(
+                os.path.join(sdir, "variants/SNVs/snv/cooccurring_mutations.csv")
+            )
         # all snv callers ('shorah', 'lofreq') produce standard VCF files
         results.append(os.path.join(sdir, "variants/SNVs/snvs.vcf"))
     # local haplotypes
     if config.output["local"]:
-        results.append(os.path.join(sdir, "variants/SNVs/snvs.csv"))
+        if config.general["snv_caller"] == "shorah":
+            results.append(os.path.join(sdir, "variants/SNVs/snvs.csv"))
+        elif config.general["snv_caller"] == "viloca":
+            results.append(
+                os.path.join(sdir, "variants/SNVs/snv/cooccurring_mutations.csv")
+            )
     # global haplotypes
     if config.output["global"]:
         if config.general["haplotype_reconstruction"] == "savage":

workflow/rules/signatures.smk

@@ -320,6 +320,9 @@ rule deconvolution:
             if config.deconvolution["variants_dates"]
             else []
         ),
+        filters=(
+            config.deconvolution["filters"] if config.deconvolution["filters"] else []
+        ),
     output:
         deconvoluted=cohortdir("deconvoluted.tsv.zst"),
         deconv_json=cohortdir("deconvoluted_upload.json"),
@@ -343,7 +346,7 @@ rule deconvolution:
     threads: config.deconvolution["threads"]
     shell:
         """
-        {params.LOLLIPOP} deconvolute "--output={output.deconvoluted}" "--out-json={output.deconv_json}" "--var={input.var_conf}" "--vd={input.var_dates}" "--dec={input.deconv_conf}" {params.out_format} {params.seed} "{input.tallymut}" 2> >(tee -a {log.errfile} >&2) > >(tee -a {log.outfile})
+        {params.LOLLIPOP} deconvolute "--output={output.deconvoluted}" "--out-json={output.deconv_json}" "--var={input.var_conf}" "--vd={input.var_dates}" "--dec={input.deconv_conf}" "--filters={input.filters}" {params.out_format} {params.seed} "{input.tallymut}" 2> >(tee -a {log.errfile} >&2) > >(tee -a {log.outfile})
         """
 
 

workflow/rules/snv.smk

@@ -377,7 +377,7 @@ rule viloca:
 
         # Get absolute path for input files
         CWD=${{PWD}}
-        WORK_DIR="$(realpath -m {ouput.WORK_DIR})"
+        WORK_DIR="$(realpath -m {output.WORK_DIR})"
         BAM="$(realpath {input.BAM})"
         REF="$(realpath {input.REF})"
         OUTFILE="$(realpath -m {log.outfile})"

workflow/schemas/config_schema.json

@@ -1661,6 +1661,12 @@
                     "default": "lines",
                     "description": "Format of the output CSV.\n- `lines`(default) - each variants a separate entry on a separate line.\n- `columns` - one column per variant",
                     "examples": ["columns"]
+                },
+                "filters": {
+                    "type": "string",
+                    "default": "",
+                    "description": "List of filters for removing problematic mutations from tally. Some mutations might be problematic and need to be taken out -- e.g. due to drop-outs in the multiplex PCR amplification, they do not show up in the data and this could be misinterpreted by LolliPop as proof of absence of a variant.",
+                    "examples": ["filters_preprint.yaml"]
                 }
             },
             "default": {},