bug fixes for REGENIE pipelines

PMBio · May 22, 2024 · 97b43ec · 97b43ec
1 parent 751c23d
commit 97b43ec
Show file tree

Hide file tree

Showing 11 changed files with 1,109 additions and 4 deletions.
diff --git a/example/annotations/preprocessing_workdir/norm/variants/variants.tsv.gz b/example/annotations/preprocessing_workdir/norm/variants/variants.tsv.gz
diff --git a/example/gencode.v38.basic.annotation.gtf.gz b/example/gencode.v38.basic.annotation.gtf.gz
diff --git a/example/genotypes.h5 b/example/genotypes.h5
diff --git a/example/imputation.bgen b/example/imputation.bgen
diff --git a/example/imputation.bgen.bgi b/example/imputation.bgen.bgi
diff --git a/example/imputation.sample b/example/imputation.sample
diff --git a/example/imputation.snplist b/example/imputation.snplist
@@ -0,0 +1,100 @@
+var0
+var1
+var2
+var3
+var4
+var5
+var6
+var7
+var8
+var9
+var10
+var11
+var12
+var13
+var14
+var15
+var16
+var17
+var18
+var19
+var20
+var21
+var22
+var23
+var24
+var25
+var26
+var27
+var28
+var29
+var30
+var31
+var32
+var33
+var34
+var35
+var36
+var37
+var38
+var39
+var40
+var41
+var42
+var43
+var44
+var45
+var46
+var47
+var48
+var49
+var50
+var51
+var52
+var53
+var54
+var55
+var56
+var57
+var58
+var59
+var60
+var61
+var62
+var63
+var64
+var65
+var66
+var67
+var68
+var69
+var70
+var71
+var72
+var73
+var74
+var75
+var76
+var77
+var78
+var79
+var80
+var81
+var82
+var83
+var84
+var85
+var86
+var87
+var88
+var89
+var90
+var91
+var92
+var93
+var94
+var95
+var96
+var97
+var98
+var99
diff --git a/example/preprocess/data/vcf/test_vcf_data_c21_b1.vcf.gz b/example/preprocess/data/vcf/test_vcf_data_c21_b1.vcf.gz
diff --git a/example/preprocess/data/vcf/test_vcf_data_c22_b1.vcf.gz b/example/preprocess/data/vcf/test_vcf_data_c22_b1.vcf.gz
diff --git a/pipelines/association_testing/regress_eval_regenie.snakefile b/pipelines/association_testing/regress_eval_regenie.snakefile
@@ -19,7 +19,10 @@ regenie_step2_bsize = regenie_config_step2["bsize"]
 regenie_njobs = regenie_config_step1.get("njobs", 1)
 regenie_joblist = range(1, regenie_njobs)
 
-
+cv_exp = config.get("cv_exp", False)
+config_file_prefix = (
+    "cv_split0/deeprvat/" if cv_exp else ""
+)
 
 wildcard_constraints:
     job="\d+"

diff --git a/pipelines/association_testing_precomputed_burdens_regenie.snakefile b/pipelines/association_testing_precomputed_burdens_regenie.snakefile
@@ -7,7 +7,7 @@ phenotypes = config['phenotypes']
 phenotypes = list(phenotypes.keys()) if type(phenotypes) == dict else phenotypes
 
 n_burden_chunks = config.get('n_burden_chunks', 1) if not debug_flag else 2
-n_regression_chunks = config.get('n_regression_chunks', 40) if not debug_flag else 2
+n_avg_chunks = config.get('n_avg_chunks', 40)
 n_bags = config['training']['n_bags'] if not debug_flag else 3
 n_repeats = config['n_repeats']
 debug = '--debug ' if debug_flag else ''
@@ -19,9 +19,9 @@ wildcard_constraints:
     trial="\d+",
 
 include: "association_testing/config.snakefile"
-include: "association_testing/association_dataset_precomputed_burdens.snakefile"
+include: "association_testing/association_dataset.snakefile"
+include: "association_testing/burdens.snakefile"
 include: "association_testing/regress_eval_regenie.snakefile"
-include: "association_testing/eval.snakefile"
 
 rule all:
     input: