Accept only-one-sample per project as input to QuaC

.test/configs/project.ped → .test/configs/project_1_sample.ped

 #family_id	sample_id	paternal_id	maternal_id	sex	phenotype
-unknown	A	-9	-9	-9	-9
-unknown	B	-9	-9	-9	-9
+unknown	A	father_1	mother_1	-9	-9

.test/configs/project_2_samples.ped

+#family_id	sample_id	paternal_id	maternal_id	sex	phenotype
+unknown	A	father_1	mother_1	-9	-9
+unknown	B	father_1	mother_1	-9	-9

Changelog.md

@@ -30,3 +30,8 @@ YYYY-MM-DD  John Doe
 * QuaC's input are output from the small variant caller pipeline, and former's output will fit seemlessly with the latter's output.
 * QuaC accepts pedigree file as input. A dummy pedigree file creator script is provided, which will be handy until phenotips is made available to us.
 * System-level testing is added
+
+2021-06-08  Manavalan Gajapathy
+
+* Bugfix: Fixes error when there is only one sample in input ped file (#34)
+* Adds system-testing for such only-one-sample-in-input setup (#35).

README.md

@@ -318,7 +318,7 @@ conda activate quac
 python src/run_quac.py \
       --project_name test_project \
       --projects_path ".test/ngs-data/" \
-      --pedigree ".test/configs/project.ped" \
+      --pedigree ".test/configs/project_2_samples.ped" \
       --outdir "$USER_SCRATCH/tmp/quac/results/test_project_wgs/analysis" \
       -e="--conda-create-envs-only"
 ```
@@ -402,27 +402,30 @@ module load Anaconda3/2020.02
 conda activate quac
 
 # WGS mode
+PROJECT="project_2_samples"
 python src/run_quac.py \
       --project_name test_project \
       --projects_path ".test/ngs-data/" \
-      --pedigree ".test/configs/project.ped" \
-      --outdir "$USER_SCRATCH/tmp/quac/results/test_project_wgs/analysis"
+      --pedigree ".test/configs/${PROJECT}.ped" \
+      --outdir "$USER_SCRATCH/tmp/quac/results/test_${PROJECT}_wgs/analysis"
 
 # Exome mode
 python src/run_quac.py \
       --project_name test_project \
       --projects_path ".test/ngs-data/" \
-      --pedigree ".test/configs/project.ped" \
-      --outdir "$USER_SCRATCH/tmp/quac/results/test_project_exome/analysis" \
+      --pedigree ".test/configs/${PROJECT}.ped" \
+      --outdir "$USER_SCRATCH/tmp/quac/results/test_${PROJECT}_exome/analysis" \
       --quac_watch_config "configs/quac_watch/exome_quac_watch_config.yaml" \
       --exome
 ```
 
+Note: Use `PROJECT="project_1_sample"` to test out a project with only one sample.
+
 ### Expected output files
 
 ```sh
-$ tree $USER_SCRATCH/tmp/quac/results/test_project/ -d -L 4
-/data/scratch/manag/tmp/quac/results/test_project/
+$ tree $USER_SCRATCH/tmp/quac/results/test_project_2_samples/ -d -L 4
+/data/scratch/manag/tmp/quac/results/test_project_2_samples/
 └── analysis
     ├── A
     │   └── qc
@@ -494,14 +497,14 @@ DAG_DIR="pipeline_visualized"
 python src/run_quac.py \
       --project_name test_project \
       --projects_path .test/ngs-data/ \
-      --pedigree .test/configs/project.ped \
+      --pedigree .test/configs/project_2_samples.ped \
       --run_locally --extra_args "--dag -F | dot -Tpng > ${DAG_DIR}/wgs_dag.png"
 
 # Rulegraph - less informative than DAG at sample level but less dense than DAG makes this easier to skim
 python src/run_quac.py \
       --project_name test_project \
       --projects_path .test/ngs-data/ \
-      --pedigree .test/configs/project.ped \
+      --pedigree .test/configs/project_2_samples.ped \
       --run_locally --extra_args "--rulegraph -F | dot -Tpng > ${DAG_DIR}/wgs_rulegraph.png"
 
 ###### Exome mode ######
@@ -509,7 +512,7 @@ python src/run_quac.py \
 python src/run_quac.py \
       --project_name test_project \
       --projects_path .test/ngs-data/ \
-      --pedigree .test/configs/project.ped \
+      --pedigree .test/configs/project_2_samples.ped \
       --exome \
       --quac_watch_config "configs/quac_watch/exome_quac_watch_config.yaml" \
       --run_locally --extra_args "--dag -F | dot -Tpng > ${DAG_DIR}/exome_dag.png"
@@ -518,7 +521,7 @@ python src/run_quac.py \
 python src/run_quac.py \
       --project_name test_project \
       --projects_path .test/ngs-data/ \
-      --pedigree .test/configs/project.ped \
+      --pedigree .test/configs/project_2_samples.ped \
       --exome \
       --quac_watch_config "configs/quac_watch/exome_quac_watch_config.yaml" \
       --run_locally --extra_args "--rulegraph -F | dot -Tpng > ${DAG_DIR}/exome_rulegraph.png"

workflow/rules/coverage_analysis.smk

@@ -124,13 +124,11 @@ rule mosdepth_plot:
         protected(OUT_DIR / "project_level_qc" / "mosdepth" / "mosdepth.html"),
     message:
         "Running mosdepth plotting"
-    params:
-        infiles=lambda wildcards: str(OUT_DIR / f"{{{','.join(SAMPLES)}}}" / "qc" / "mosdepth" / "*.mosdepth.global.dist.txt"),
     shell:
         r"""
         python {input.script} \
             --output {output} \
-            {params.infiles}
+            {input.dist}
         """
 
 
@@ -156,12 +154,11 @@ rule indexcov:
         str(WORKFLOW_PATH / "configs/env/goleft.yaml")
     params:
         outdir=lambda wildcards, output: Path(output[0]).parent,
-        infiles=lambda wildcards: str(PROJECT_PATH / f"{{{','.join(SAMPLES)}}}" / "bam" / "*.bam"),
     shell:
         r"""
         goleft indexcov \
             --directory {params.outdir} \
-            {params.infiles} \
+            {input.bam} \
             > {log} 2>&1
         """
 

workflow/rules/relatedness_ancestry.smk

@@ -41,14 +41,13 @@ rule somalier_relate:
         log=OUT_DIR / "project_level_qc" / "somalier" / "relatedness" / "somalier.log",
     params:
         outdir=lambda wildcards, output: Path(output["out"][0]).parent,
-        infiles=lambda wildcards: str(OUT_DIR / "project_level_qc" / "somalier" / "extract" / f"{{{','.join(SAMPLES)}}}.somalier"),
     shell:
         r"""
         somalier relate \
             --ped {input.ped} \
             --infer \
             --output-prefix {params.outdir}/somalier \
-            {params.infiles} \
+            {input.extracted} \
             > {log} 2>&1
         """
 
@@ -73,13 +72,12 @@ rule somalier_ancestry:
         log=OUT_DIR / "project_level_qc" / "somalier" / "ancestry" / "somalier.log",
     params:
         outdir=lambda wildcards, output: Path(output["out"][0]).parent,
-        infiles=lambda wildcards: str(OUT_DIR / "project_level_qc" / "somalier" / "extract" / f"{{{','.join(SAMPLES)}}}.somalier"),
     shell:
         r"""
         somalier ancestry \
             --output-prefix {params.outdir}/somalier \
             --labels {input.labels_1kg} \
             {input.somalier_1kg_directory}/*.somalier ++ \
-            {params.infiles} \
+            {input.extracted} \
             > {log} 2>&1
         """