cwhelan

	workspace:cloud_sdk_docker	cnmops_docker	condense_counts_docker	gatk_docker	gatk_docker_pesr_override	genomes_in_the_cloud_docker	linux_docker	manta_docker	samtools_cloud_docker	sv_base_docker	sv_base_mini_docker	sv_pipeline_base_docker	sv_pipeline_docker	sv_pipeline_hail_docker	sv_pipeline_updates_docker	sv_pipeline_qc_docker	sv_pipeline_rdtest_docker	wham_docker	ref_panel_name	ref_panel_bincov_matrix	ref_panel_contig_ploidy_model_tar	ref_panel_cutoffs	ref_panel_del_bed	ref_panel_dup_bed	ref_panel_gcnv_model_tars_list	ref_panel_genotype_pesr_pesr_sepcutoff	ref_panel_genotype_pesr_depth_sepcutoff	ref_panel_genotype_depth_pesr_sepcutoff	ref_panel_genotype_depth_depth_sepcutoff	ref_panel_ped_file	ref_panel_PE_files_list	ref_panel_PE_metrics	ref_panel_qc_definitions	ref_panel_requester_pays_crams	ref_panel_samples_list	ref_panel_SD_files_list	ref_panel_SR_files_list	ref_panel_SR_metrics	ref_panel_std_manta_vcf_tar	ref_panel_std_wham_vcf_tar	ref_panel_vcf	reference_name	reference_allosome_file	reference_autosome_

	phaseSets <- read.table("phase_sets.dat", header=T)
	switches <- read.table("compare.err_pos.more") // this file is created by the phasing evaluation scripts
	names(switches) <- c("POS", "TYPE")

	phaseSets$MID <- (phaseSets$START + phaseSets$END) / 2
	phaseSets$COLOR <- factor(seq(1,dim(phaseSets)[1]) %% 4)
	phaseSets$LABEL <- ifelse(phaseSets$NUM_SITES > 500, phaseSets$NUM_SITES, NA)

	ggplot(phaseSets) +
	geom_segment(aes(y=1, yend=1, x=START, xend=END, color=COLOR), size=4) +

	library(GenomicRanges)
	library(rtracklayer)
	library(bsseq)
	library(Homo.sapiens)

	# methylation summaries are data frames with columns, "chr", "start", "strand", "meth", "unmeth"

	# get them all in one data frame
	mergedSummary <- merge(normalMethylationSummary, cancerMethylationSummary,
	by=c("chr", "start", "strand", "end"),