Fixes of main worklflow

Author: aleksandr-medvedev-msc

config.yaml

@@ -2,7 +2,7 @@ mode: all
 samples_file: samples.tsv
 vcf_file: vcf/merged.vcf.gz
 reference:
-  ref_dir: /media_ssd/ref
+  ref_dir: /media/ref
   GRCh37_fasta: human_g1k_v37.fasta
   GENETIC_MAP: tables/genetic_map_hg19_withX.txt.gz
   vcfRef: 1000genome/bcf/1000genome_chr{chrom}.bcf

rules/relatives.smk

@@ -124,11 +124,11 @@ rule merge_matches:
 rule merge_ibd_segments:
     input:
         germline=rules.merge_matches.output[0],
-        true_ibd=rules.simulate.output.seg
     params:
         cm_dir='cm',
         merge_gap='0.6',
-        use_true_ibd=use_simulated_ibd
+        use_true_ibd=use_simulated_ibd,
+        true_ibd='pedsim/simulated/data.seg' # it is in the params because in the case of true data we do not have this information
     output:
         ibd='ibd/merged_ibd.tsv'
     conda: "../envs/evaluation.yaml"

scripts/merge_ibd.py

@@ -6,7 +6,7 @@
     germline_path = snakemake.input['germline']
     map_dir = snakemake.params['cm_dir']
     gap = float(snakemake.params['merge_gap'])
-    pedsim_path = snakemake.input['true_ibd']
+    pedsim_path = snakemake.params['true_ibd']
 
     use_true_ibd = bool(snakemake.params['use_true_ibd'])
     if not use_true_ibd: