fix handling of indels and sample names

src/harpy/phase.py

@@ -78,19 +78,20 @@ def phase(inputs, output_dir, vcf, threads, molecule_distance, prune_threshold,
     with open(f"{workflowdir}/config.yaml", "w", encoding="utf-8") as config:
         config.write("workflow: phase\n")
         config.write(f"output_directory: {output_dir}\n")
-        config.write(f"noBX: {ignore_bx}\n")
+        config.write(f"ignore_bx: {ignore_bx}\n")
         config.write(f"prune: {prune_threshold/100}\n")
         config.write(f"molecule_distance: {molecule_distance}\n")
+        config.write(f"samples_from_vcf: {vcf_samples}\n")
         if extra_params is not None:
             config.write(f"extra: {extra_params}\n")
-        config.write(f"skipreports: {skipreports}\n")
+        config.write(f"skip_reports: {skipreports}\n")
         config.write(f"workflow_call: {command}\n")
         config.write("inputs:\n")
         config.write(f"  variantfile: {vcf}\n")
         if genome is not None:
             config.write(f"  genome: {genome}\n")
-            if not os.path.exists(f"{genome}.fai"):
-                subprocess.run(f"samtools faidx --fai-idx {genome}.fai {genome}".split())
+        #    if not os.path.exists(f"{genome}.fai"):
+        #        subprocess.run(f"samtools faidx --fai-idx {genome}.fai {genome}".split())
         config.write("  alignments:\n")
         for i in bamlist:
             config.write(f"    - {i}\n")

src/harpy/snakefiles/phase.smk

@@ -1,6 +1,7 @@
 containerized: "docker://pdimens/harpy:latest"
 
 import sys
+import subprocess
 import multiprocessing
 from pathlib import Path
 from rich import print as rprint
@@ -13,20 +14,27 @@ pruning           = config["prune"]
 molecule_distance = config["molecule_distance"]
 extra             = config.get("extra", "") 
 outdir 			  = config["output_directory"]
-envdir      = os.getcwd() + "/.harpy_envs"
-skipreports = config["skipreports"]
-
+envdir            = os.getcwd() + "/.harpy_envs"
+skipreports       = config["skip_reports"]
+samples_from_vcf  = config["samples_from_vcf"]
 variantfile       = config["inputs"]["variantfile"]
 bamlist     = config["inputs"]["alignments"]
-samplenames = [Path(i).stem for i in bamlist]
+
 # toggle linked-read aware mode
-if config["noBX"]:
+if config["ignore_bx"]:
     fragfile = outdir + "/extractHairs/{sample}.unlinked.frags"
     linkarg = "--10x 0"
 else:
     fragfile =  outdir + "/linkFragments/{sample}.linked.frags"
     linkarg  = "--10x 1"
 
+if samples_from_vcf:
+    bcfquery = subprocess.run(["bcftools", "query", "-l", variantfile], stdout=subprocess.PIPE)
+    samplenames = bcfquery.stdout.read().decode().split()
+else:
+    samplenames = [Path(i).stem for i in bamlist]
+
+
 # toggle indel mode
 if config["inputs"].get("genome", None):
     genomefile = config["inputs"]["genome"]
@@ -41,7 +49,7 @@ else:
     geno       = []
     genofai    = []
     bn         = []
-    indels     = True
+    indels     = False
 
 wildcard_constraints:
     sample = "[a-zA-Z0-9._-]+"