novoCaller triofiles order change, comHet can deal with more than 3 sample in trio argument

Author: Michele Berselli

README.md

@@ -11,7 +11,7 @@ For more details, see granite [*documentation*](https://granite-suite.readthedoc
 ## Availability and requirements
 A ready-to-use docker image is available to download.
 
-    docker pull b3rse/granite:v0.1.11
+    docker pull b3rse/granite:v0.1.12
 
 To run locally, install the following libraries:
 
@@ -156,3 +156,7 @@ rckTar creates a tar archive from bgzip and tabix indexed RCK files. Creates an
 
 ### validateVCF
 validateVCF allows to calculate error models for different inheritance modes for input VCF file using pedigree information.
+
+## WARNINGS
+
+!! starting from 0.1.12, novoCaller `--triofiles` expected order changed. Now PROBAND must be listed as first.

docs/callers.md

@@ -22,7 +22,7 @@ novoCaller is a Bayesian calling algorithm for *de novo* mutations. The model us
                             bgzip and tabix indexed RCK)
       -t TRIOFILES, --triofiles TRIOFILES
                             TSV index file containing SampleID<TAB>Path/to/file
-                            for family files, the PROBAND must be listed as LAST
+                            for family files, the PROBAND must be listed as FIRST
                             (BAM or bgzip and tabix indexed RCK)
       --ppthr PPTHR         threshold to filter by posterior probabilty for de
                             novo calls (>=) [0]

docs/conf.py

@@ -22,7 +22,7 @@
 author = 'Michele Berselli'
 
 # The full version, including alpha/beta/rc tags
-release = '0.1.11'
+release = '0.1.12'
 
 
 # -- General configuration ---------------------------------------------------

docs/install.md

@@ -2,7 +2,7 @@
 
 A ready-to-use docker image is available to download.
 
-    docker pull b3rse/granite:v0.1.11
+    docker pull b3rse/granite:v0.1.12
 
 To run locally, Python 3.6+ is required together with the following libraries:
 

granite/_version.py

@@ -1,4 +1,4 @@
 """Version information."""
 
 # The following line *must* be the last in the module, exactly as formatted:
-__version__ = "0.1.11"
+__version__ = "0.1.12"

granite/comHet.py

@@ -616,12 +616,12 @@ def main(args, test=False):
     #end if
 
     # Get trio IDs
-    if len(args['trio']) > 3:
-        sys.exit('\nERROR in parsing arguments: too many sample IDs provided for trio\n')
-    elif len(args['trio']) == 3:
+    # if len(args['trio']) > 3:
+    #     sys.exit('\nERROR in parsing arguments: too many sample IDs provided for trio\n')
+    if len(args['trio']) == 3:
         is_family = True
     #end if
-    ID_list = args['trio'] # [proband_ID, parent_ID, parent_ID]
+    ID_list = args['trio'][:3] # [proband_ID, parent_ID, parent_ID]
 
     # Reading variants
     analyzed = 0

granite/granite.py

@@ -56,7 +56,7 @@ def main():
     novoCaller_parser.add_argument('-i', '--inputfile', help='input VCF file', type=str, required=True)
     novoCaller_parser.add_argument('-o', '--outputfile', help='output file to write results as VCF, use .vcf as extension', type=str, required=True)
     novoCaller_parser.add_argument('-u', '--unrelatedfiles', help='TSV index file containing SampleID<TAB>Path/to/file for unrelated files used to train the model (BAM or bgzip and tabix indexed RCK)', type=str, required=True)
-    novoCaller_parser.add_argument('-t', '--triofiles', help='TSV index file containing SampleID<TAB>Path/to/file for family files, the PROBAND must be listed as LAST (BAM or bgzip and tabix indexed RCK)', type=str, required=True)
+    novoCaller_parser.add_argument('-t', '--triofiles', help='TSV index file containing SampleID<TAB>Path/to/file for family files, the PROBAND must be listed as FIRST (BAM or bgzip and tabix indexed RCK)', type=str, required=True)
     novoCaller_parser.add_argument('--ppthr', help='threshold to filter by posterior probabilty for de novo calls (>=) [0]', type=float, required=False)
     novoCaller_parser.add_argument('--afthr', help='threshold to filter by population allele frequency (<=) [1]', type=float, required=False)
     novoCaller_parser.add_argument('--afthr_unrelated', help='threshold to filter by allele frequency calculated among unrelated (<=) [1]', type=float, required=False)

granite/novoCaller.py

@@ -843,12 +843,24 @@ def main(args, test=False):
 
     if is_bam: # if bam files
         unrelated_files, IDs_unrelated = buffering_bams(args['unrelatedfiles'])
-        trio_files, IDs_trio = buffering_bams(args['triofiles']) # [parent, parent, child]
+        trio_files, IDs_trio = buffering_bams(args['triofiles'])
     else:
         unrelated_files, IDs_unrelated = buffering_rcks(args['unrelatedfiles'])
-        trio_files, IDs_trio = buffering_rcks(args['triofiles']) # [parent, parent, child]
+        trio_files, IDs_trio = buffering_rcks(args['triofiles'])
     #end if
 
+    # !!!
+    # Subset trio_files and IDs_trio to exclude additional family members
+    trio_files = trio_files[:3]
+    IDs_trio = IDs_trio[:3]
+    # Sorting trio_files and IDs_trio to have proband last
+    #   [child, parent, parent] this is as it is expected in the input file
+    #   the code expect instead [parent, parent, child]
+    #   we need to put child at the end
+    trio_files.append(trio_files.pop(0))
+    IDs_trio.append(IDs_trio.pop(0))
+    # !!!
+
     # Checking info files for trio is complete
     if len(trio_files) != 3:
         sys.exit('\nERROR in BAMs info file for trio: missing information for some family member\n')

tests/files/trio_bam.tsv

@@ -1,3 +1,4 @@
+HG002	tests/files/bam/HG002.hs37d5.60x.1.addrg_sorted.bam_regions.bam
 HG003	tests/files/bam/HG003.hs37d5.60x.1.addrg_sorted.bam_regions.bam
 HG004	tests/files/bam/HG004.hs37d5.60x.1.addrg_sorted.bam_regions.bam
-HG002	tests/files/bam/HG002.hs37d5.60x.1.addrg_sorted.bam_regions.bam
+HG003	tests/files/bam/HG003.hs37d5.60x.1.addrg_sorted.bam_regions.bam

tests/files/trio_rck.tsv

@@ -1,3 +1,3 @@
+UDN930725	tests/files/rck/GAPFIZWAW2KS_mpileup.out.gz_120Kbp.rck.gz
 MOM	tests/files/rck/GAPFIXRIG4PE_mpileup.out.gz_120Kbp.rck.gz
 DAD	tests/files/rck/GAPFIYDBF875_mpileup.out.gz_120Kbp.rck.gz
-UDN930725	tests/files/rck/GAPFIZWAW2KS_mpileup.out.gz_120Kbp.rck.gz