Customize battenberg_wgs.R wrapper

CHANGELOG.md

@@ -6,8 +6,14 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).
 This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 ---
+## Unreleased
+### Added
+- Add custom `battenberg_wgs.R`
+
+### Removed
+- Remove `modify_reference_path.sh`
 
-## [2.2.9] - 2023-06-27
+## [2.2.9] - 2023-06-27 [YANKED]
 ### Added
 - Add `modify_reference_path.sh`
 - Add GRCh37 and GRCh38 resource paths to `README.md`

Dockerfile

@@ -30,14 +30,14 @@
         "gridExtra","doParallel","foreach", "splines", "VariantAnnotation", "copynumber"))'
 
 # Install devtools, ASCAT & Battenberg
+FROM r-base:latest
 RUN R -q -e 'install.packages("devtools", dependencies = TRUE)' && \
     R -q -e 'devtools::install_github("Crick-CancerGenomics/ascat/[email protected]")' && \
     R -q -e 'devtools::install_github("Wedge-Oxford/[email protected]")'
 
-# Modify paths to reference files
-COPY modify_reference_path.sh /usr/local/bin/modify_reference_path.sh
-RUN chmod +x /usr/local/bin/modify_reference_path.sh && \
-    bash /usr/local/bin/modify_reference_path.sh /usr/local/lib/R/site-library/Battenberg/example/battenberg_wgs.R /usr/local/bin/battenberg_wgs.R
+# Add custom Battenberg R wrapper
+COPY battenberg_wgs.R /usr/local/bin/battenberg_wgs.R
+RUN chmod +x /usr/local/bin/battenberg_wgs.R
 
 RUN ln -sf /usr/local/lib/R/site-library/Battenberg/example/filter_sv_brass.R /usr/local/bin/filter_sv_brass.R && \
     ln -sf /usr/local/lib/R/site-library/Battenberg/example/battenberg_cleanup.sh /usr/local/bin/battenberg_cleanup.sh

battenberg_wgs.R

@@ -0,0 +1,121 @@
+###############################################################################
+# A pure R Battenberg v2.2.9 WGS pipeline implementation.
+###############################################################################
+library(Battenberg);
+library(optparse);
+
+option.list <- list(
+    make_option(c('-t', '--tumourname'), type = 'character', default = NULL, help = 'Samplename of the tumour', metavar = 'character'),
+    make_option(c('-n', '--normalname'), type = 'character', default = NULL, help = 'Samplename of the normal', metavar = 'character'),
+    make_option(c('--tb'), type = 'character', default = NULL, help = 'Tumour BAM file', metavar = 'character'),
+    make_option(c('--nb'), type = 'character', default = NULL, help = 'Normal BAM file', metavar = 'character'),
+    make_option(c('--sex'), type = 'character', default = NULL, help = 'Sex of the sample', metavar = 'character'),
+    make_option(c('-o', '--output'), type = 'character', default = NULL, help = 'Directory where output will be written', metavar = 'character'),
+    make_option(c('--skip_allelecount'), type = 'logical', default = FALSE, action = 'store_true', help = 'Provide when alleles don\'t have to be counted. This expects allelecount files on disk', metavar = 'character'),
+    make_option(c('--skip_preprocessing'), type = 'logical', default = FALSE, action = 'store_true', help = 'Provide when pre-processing has previously completed. This expects the files on disk', metavar = 'character'),
+    make_option(c('--skip_phasing'), type = 'logical', default = FALSE, action = 'store_true', help = 'Provide when phasing has previously completed. This expects the files on disk', metavar = 'character'),
+    make_option(c('--cpu'), type = 'numeric', default = 8, help = 'The number of CPU cores to be used by the pipeline (Default: 8)', metavar = 'character'),
+    make_option(c('--bp'), type = 'character', default = NULL, help = 'Optional two column file (chromosome and position) specifying prior breakpoints to be used during segmentation', metavar = 'character'),
+    make_option(c('--min_ploidy'), type = 'double', default = 1.6, help = 'The minimum ploidy to consider', metavar = 'character'),
+    make_option(c('--max_ploidy'), type = 'double', default = 4.8, help = 'The maximum ploidy to consider', metavar = 'character'),
+    make_option(c('--min_rho'), type = 'double', default = 0.1, help = 'The minimum cellularity to consider', metavar = 'character'),
+    make_option(c('--platform_gamma'), type = 'numeric', default = 1, help = 'Platform specific gamma value (0.55 for SNP6, 1 for NGS)', metavar = 'character'),
+    make_option(c('--phasing_gamma'), type = 'numeric', default = 1, help = 'Gamma parameter used when correcting phasing mistakes (Default: 1)', metavar = 'character'),
+    make_option(c('--segmentation_gamma'), type = 'numeric', default = 10, help = 'The gamma parameter controls the size of the penalty of starting a new segment during segmentation. It is therefore the key parameter for controlling the number of segments (Default: 10)', metavar = 'character'),
+    make_option(c('--segmentation_kmin'), type = 'numeric', default = 3, help = 'Kmin represents the minimum number of probes/SNPs that a segment should consist of (Default: 3)', metavar = 'character'),
+    make_option(c('--phasing_kmin'), type = 'numeric', default = 1, help = 'Kmin used when correcting for phasing mistakes (Default: 3)', metavar = 'character'),
+    make_option(c('--clonality_dist_metric'), type = 'numeric', default = 0, help = 'Distance metric to use when choosing purity/ploidy combinations (Default: 0)', metavar = 'character'),
+    make_option(c('--ascat_dist_metric'), type = 'numeric', default = 1, help = 'Distance metric to use when choosing purity/ploidy combinations (Default: 1)', metavar = 'character'),
+    make_option(c('--min_goodness_of_fit'), type = 'double', default = 0.63, help = 'Minimum goodness of fit required for a purity/ploidy combination to be accepted as a solution (Default: 0.63)', metavar = 'character'),
+    make_option(c('--balanced_threshold'), type = 'double', default = 0.51, help = 'The threshold beyond which BAF becomes uninformative (Default: 0.51)', metavar = 'character'),
+    make_option(c('--min_normal_depth'), type = 'numeric', default = 10, help = 'Minimum depth required in the matched normal for a SNP to be considered as part of the wgs analysis (Default: 10)', metavar = 'character'),
+    make_option(c('--min_base_qual'), type = 'numeric', default = 20, help = 'Minimum base quality required for a read to be counted when allele counting (Default: 20)', metavar = 'character'),
+    make_option(c('--min_map_qual'), type = 'numeric', default = 35, help = 'Minimum mapping quality required for a read to be counted when allele counting (Default: 35)', metavar = 'character'),
+    make_option(c('--calc_seg_baf_option'), type = 'numeric', default = 3, help = 'Sets way to calculate BAF per segment: 1=mean, 2=median, 3=ifelse median==0 | 1, mean, median (Default: 3)', metavar = 'character'),
+    make_option(c('--data_type'), type = 'character', default = 'wgs', help = 'String that contains either wgs or snp6 depending on the supplied input data (Default: wgs)', metavar = 'character')
+    );
+
+opt.parser <- OptionParser(option_list = option.list);
+opt <- parse_args(opt.parser);
+
+TUMOURNAME <- opt$tumourname;
+NORMALNAME <- opt$normalname;
+NORMALBAM <- opt$nb;
+TUMOURBAM <- opt$tb;
+IS.MALE <- opt$sex == 'male' | opt$sex == 'Male';
+RUN.DIR <- opt$output;
+SKIP.ALLELECOUNTING <- opt$skip_allelecount;
+SKIP.PREPROCESSING <- opt$skip_preprocessing;
+SKIP.PHASING <- opt$skip_phasing;
+NTHREADS <- opt$cpu;
+PRIOR.BREAKPOINTS.FILE <- opt$bp;
+MIN.PLOIDY <- opt$min_ploidy;
+MAX.PLOIDY <- opt$max_ploidy;
+MIN.RHO <- opt$min_rho;
+PLATFORM.GAMMA <- opt$platform_gamma;
+PHASING.GAMMA <- opt$phasing_gamma;
+SEGMENTATION.GAMMA <- opt$segmentation_gamma;
+SEGMENTATION.KMIN <- opt$segmentation_kmin;
+PHASING.KMIN <- opt$phasing_kmin;
+CLONALITY.DIST.METRIC <- opt$clonality_dist_metric;
+ASCAT.DIST.METRIC <- opt$ascat_dist_metric;
+MIN.GOODNESS.OF.FIT <- opt$min_goodness_of_fit;
+BALANCED.THRESHOLD <- opt$balanced_threshold;
+MIN.NORMAL.DEPTH <- opt$min_normal_depth;
+MIN.BASE.QUAL <- opt$min_base_qual;
+MIN.MAP.QUAL <- opt$min_map_qual;
+CALC.SEG.BAF.OPTION <- opt$calc_seg_baf_option;
+DATA.TYPE <- opt$data_type;
+
+# General static
+IMPUTEINFOFILE <- '/opt/battenberg_reference/impute_info.txt';
+G1000PREFIX <- '/opt/battenberg_reference/1000_genomes_loci/1000_genomes_allele_index_chr';
+G1000PREFIX.AC <- '/opt/battenberg_reference/1000_genomes_loci/1000_genomes_loci_chr';
+GCCORRECTPREFIX <- '/opt/battenberg_reference/1000_genomes_gcContent/1000_genomes_GC_corr_chr';
+REPLICCORRECTPREFIX <- '/opt/battenberg_reference/battenberg_wgs_replication_timing_correction_1000_genomes/1000_genomes_replication_timing_chr';
+IMPUTE.EXE <- 'impute2';
+
+# WGS specific static
+ALLELECOUNTER <- 'alleleCounter';
+PROBLEMLOCI <- '/opt/battenberg_reference/battenberg_problem_loci/probloci.txt.gz';
+
+# Change to work directory and load the chromosome information
+setwd(RUN.DIR);
+
+battenberg(
+    tumourname = TUMOURNAME,
+    normalname = NORMALNAME,
+    tumour_data_file = TUMOURBAM,
+    normal_data_file = NORMALBAM,
+    ismale = IS.MALE,
+    imputeinfofile = IMPUTEINFOFILE,
+    g1000prefix = G1000PREFIX,
+    g1000allelesprefix = G1000PREFIX.AC,
+    gccorrectprefix = GCCORRECTPREFIX,
+    repliccorrectprefix = REPLICCORRECTPREFIX,
+    problemloci = PROBLEMLOCI,
+    data_type = DATA.TYPE,
+    impute_exe = IMPUTE.EXE,
+    allelecounter_exe = ALLELECOUNTER,
+    nthreads = NTHREADS,
+    platform_gamma = PLATFORM.GAMMA,
+    phasing_gamma = PHASING.GAMMA,
+    segmentation_gamma = SEGMENTATION.GAMMA,
+    segmentation_kmin = SEGMENTATION.KMIN,
+    phasing_kmin = PHASING.KMIN,
+    clonality_dist_metric = CLONALITY.DIST.METRIC,
+    ascat_dist_metric = ASCAT.DIST.METRIC,
+    min_ploidy = MIN.PLOIDY,
+    max_ploidy = MAX.PLOIDY,
+    min_rho = MIN.RHO,
+    min_goodness = MIN.GOODNESS.OF.FIT,
+    uninformative_BAF_threshold = BALANCED.THRESHOLD,
+    min_normal_depth = MIN.NORMAL.DEPTH,
+    min_base_qual = MIN.BASE.QUAL,
+    min_map_qual = MIN.MAP.QUAL,
+    calc_seg_baf_option = CALC.SEG.BAF.OPTION,
+    skip_allele_counting = SKIP.ALLELECOUNTING,
+    skip_preprocessing = SKIP.PREPROCESSING,
+    skip_phasing = SKIP.PHASING,
+    prior_breakpoints_file = PRIOR.BREAKPOINTS.FILE
+    );

metadata.yaml

@@ -4,7 +4,7 @@ description: 'Docker repository for Wedge-lab/battenberg'
 maintainers: ['[email protected]']
 languages: ['Dockerfile']
 tools: ['battenberg']
-version: ['2.2.9']  # Tool version number
-purpose: 'Whole Genome Sequencing subclonal copy number caller'  # Description of what this tool does
-references: 'https://github.com/Wedge-lab/battenberg'  # is the tool/dependencies published, is there a confluence page
-image_name: 'battenberg'  # name of the new docker image
+version: ['2.2.9']
+purpose: 'Whole Genome Sequencing subclonal copy number caller'
+references: 'https://github.com/Wedge-lab/battenberg'
+image_name: 'battenberg'