atacseq_config.yaml

# The following section contains resource files required by the pipeline
adapter_fasta: /research/lab_bsf/resources/atacseq/nextera_adapters.fa
bowtie2_index:
    hg38: /nobackup/lab_bsf/resources/genomes/hg38/indices_for_Bowtie2/hg38
    mm10: /nobackup/lab_bsf/resources/genomes/mm10/indices_for_Bowtie2/mm10
chromosome_sizes:
    hg38: /research/lab_bsf/resources/atacseq/hg38.chromSizes
    mm10: /research/lab_bsf/resources/atacseq/mm10.chromSizes
blacklisted_regions:
    hg38: /research/lab_bsf/resources/atacseq/hg38-ENCODE_blacklist.sorted.v3.bed
    mm10: /research/lab_bsf/resources/atacseq/mm10-ENCODE_blacklist.sorted.v2.bed
whitelisted_regions:
    hg38: /research/lab_bsf/resources/atacseq/hg38-ENCODE_whitelist.sorted.v3.bed
    mm10: /research/lab_bsf/resources/atacseq/mm10-ENCODE_whitelist.sorted.v2.bed
unique_tss:
    hg38: /research/lab_bsf/resources/atacseq/hg38_gencode_tss_unique.sorted.bed
    mm10: /research/lab_bsf/resources/atacseq/mm10_CCDS_tss_unique.bed
regulatory_regions:
    hg38: /research/lab_bsf/resources/atacseq/homo_sapiens.GRCh38.Regulatory_Build.regulatory_features.chromosomes_only.20161111.bed
    mm10: /research/lab_bsf/resources/atacseq/mus_musculus.GRCm38.Regulatory_Build.regulatory_features.20161111.sorted.bed
genome_sizes:
    hg38: 2747877777
    mm10: 2407883318
mitochondria_names:
    hg38: chrM
    mm10: chrM


data_sources:
  # specify the ABSOLUTE PATH of input files using variable path expressions
  # entries correspond to values in the data_source column in sample_annotation table
  # {variable} can be used to replace environment variables or other sample_annotation columns
  # If you use {variable} codes, you should quote the field so python can parse it.
  bsf: /nobackup/lab_bsf/samples/{flowcell}/{flowcell}_{lane}_samples/{flowcell}_{lane}#{BSF_name}.bam
  bsf_test: /nobackup/lab_bsf/users/berguener/junk/atac_test_data/{flowcell}_{lane}#{BSF_name}.bam