fix: change on yaml and id range (#86)

bin/accession_id.sh

@@ -23,7 +23,8 @@ then
    helpFunction
 fi
 
-id=$(echo $GCST|sed 's/GCST/1/g')
+ida=$(echo $GCST|sed 's/GCST/1/g')
+id=$(($ida-1))
 lower=$((10#$id/1000*1000+1))
 upper=$(((10#$id/1000+1)*1000))
 

bin/lib/SumStatRecord.py

@@ -13,7 +13,7 @@ def __init__(self, chrom, pos, other_al, effect_al, beta, oddsr,
         self.other_al = other_al
         self.effect_al = effect_al
         self.data = data
-        self.beta = float(beta) if beta is not None else None
+        self.beta = safe_float(beta) if beta is not None else None
         self.oddsr = safe_float(oddsr) if oddsr is not None else None
         self.oddsr_lower = safe_float(oddsr_lower) if oddsr_lower is not None else None
         self.oddsr_upper = safe_float(oddsr_upper) if oddsr_upper is not None else None
@@ -74,8 +74,9 @@ def flip_beta(self):
                             to flipping.
         """
         # Flip beta
-        if self.beta and self.beta != 0:
-            self.beta = self.beta * -1
+        if self.beta:
+            if self.beta != 0:
+                self.beta = self.beta * -1
         # Flip OR
         if self.oddsr:
             self.oddsr = self.oddsr ** -1

bin/main_pysam.py

@@ -165,9 +165,12 @@ def main():
             out_raw["hm_coordinate_conversion"] = ss_rec.data["hm_coordinate_conversion"]
             out_raw["variant_id"] = vcf_rec.hgvs()[0] if vcf_rec and ss_rec.is_harmonised else args.na_rep_out
             out_raw["rsid"] = ss_rec.hm_rsid if vcf_rec and ss_rec.is_harmonised else args.na_rep_out
-            out_raw["standard_error"]=ss_rec.data["standard_error"] if ss_rec.data["standard_error"] is not None else args.na_rep_out
+            try:
+                out_raw["standard_error"]=ss_rec.data["standard_error"] if ss_rec.data["standard_error"] is not None else args.na_rep_out
+            except:
+                out_raw["standard_error"]=args.na_rep_out
             # Add other data from summary stat file
-            outed=["chromosome","base_pair_location","p_value","effect_allele","other_allele","effect_allele_frequency","beta","odds_ratio","standard_error","rsid","ci_upper","ci_lower","hm_coordinate_conversion"]
+            outed=["chromosome","base_pair_location","p_value","effect_allele","other_allele","effect_allele_frequency","beta","odds_ratio","rsid","standard_error","ci_upper","ci_lower","hm_coordinate_conversion"]
             for key in ss_rec.data:
                 if key not in outed:
                     value = ss_rec.data[key] if ss_rec.data[key] else args.na_rep_out
@@ -262,7 +265,7 @@ def parse_args():
                                   '(b) 0_base '),
                             nargs='?',
                             type=str,
-                            const="1-based")
+                            default="1-based")
     mode_group.add_argument('--palin_mode', metavar="[infer|forward|reverse|drop]",
                         help=('Mode to use for palindromic variants:\n'
                               '(a) infer strand from effect-allele freq, '

modules/local/harmonization_log.nf

@@ -35,19 +35,20 @@ process harmonization_log {
     data_file_name="${GCST}.h.tsv.gz"
     out_yaml="${GCST}.h.tsv.gz-meta.yaml"
     data_file_md5sum=\$(md5sum<${launchDir}/$GCST/final/${GCST}.h.tsv.gz | awk '{print \$1}')
-    date_last_modified=\$(date  +"%Y-%m-%d")
+    date_metadata_last_modified=\$(date  +"%Y-%m-%d")
     harmonisation_reference=\$(tabix -H "${params.ref}/homo_sapiens-${chr}.vcf.gz" | grep reference | cut -f2 -d '=')
 
     gwas_metadata.py \
     -i $raw_yaml \
     -o \$out_yaml \
-    --dataFileName \$data_file_name \
+    -e \
+    --data_file_name \$data_file_name \
     --data_file_md5sum \$data_file_md5sum \
     --is_harmonised True \
     --is_sorted True \
     --genome_assembly GRCh38 \
     --coordinate_system 1-based \
-    --date_last_modified \$date_last_modified \
+    --date_metadata_last_modified \$date_metadata_last_modified \
     --harmonisation_reference \$harmonisation_reference \
     """
 }