Skip to content

feat: add hs1_chrR as a supported reference genome #117

Closed
4 of 4 issues completed
Closed

Description

@kopardev

Summary

Add support for the hs1_chrR reference genome in ASPEN. This is a variant of the
existing hs1 (T2T CHM13v2.0) reference but restricted to canonical chromosomes only
(chrR = "chromosomes only" subset, no alt/patch/unplaced contigs). Requested by
@wuy24 for project CCRTEGS6.

Background

hs1 is already supported. hs1_chrR follows the same pattern as other _chrR-style
references and requires the same set of resource files to be built/registered.

Tasks

1. Bowtie2 index + accessory index files (Biowulf /data/CCBR_Pipeliner/db/PipeDB/Indices/)

Build or locate the existing index directory and verify the following files exist:

  • hs1_chrR/indexes/hs1_chrR.fa — genome FASTA
  • hs1_chrR/indexes/hs1_chrR.genome — chrom sizes (required by MACS2)
  • hs1_chrR/indexes/*.bt2l — Bowtie2 index files (large index format, genome > 4 GB)
  • hs1_chrR/indexes/hs1_chrR.Ns.bed.gz — N-regions bed (optional, used by Genrich -E)
  • hs1_chrR/indexes/hs1_chrR.blacklist.bed.gz — blacklist BED (optional, used by Genrich -E)

2. Pipeline resource files (under resources/)

Create and add the following files (follow the hs1 equivalents as templates):

  • resources/chroms/hs1_chrR.chroms — space-separated list of canonical chromosomes
  • resources/blacklistFa/hs1_chrR.blacklist.fa.gz — blacklist FASTA (LFS-tracked)
  • resources/tssBed/hs1_chrR_tssbeds.tar.gz — per-gene TSS bins (LFS-tracked); reused from hs1_tssbeds.tar.gz
  • resources/frip/hs1_chrR.DHS.bed.gz — DHS regions (LFS-tracked); reused from hs1.DHS.bed.gz
  • resources/frip/hs1_chrR.promoters.bed.gz — promoter regions (LFS-tracked); reused from hs1.promoters.bed.gz
  • resources/frip/hs1_chrR.enhancers.bed.gz — enhancer regions (LFS-tracked); reused from hs1.enhancers.bed.gz

3. Register LFS tracking (.gitattributes)

Add LFS filter lines for all new binary/gz resource files, mirroring the hs1 entries:

resources/blacklistFa/hs1_chrR.blacklist.fa.gz filter=lfs diff=lfs merge=lfs -text
resources/frip/hs1_chrR.enhancers.bed.gz filter=lfs diff=lfs merge=lfs -text
resources/tssBed/hs1_chrR_tssbeds.tar.gz filter=lfs diff=lfs merge=lfs -text
resources/frip/hs1_chrR.DHS.bed.gz filter=lfs diff=lfs merge=lfs -text
resources/frip/hs1_chrR.promoters.bed.gz filter=lfs diff=lfs merge=lfs -text

✅ All entries added to .gitattributes and committed.

4. config/config.yaml

Add the hs1_chrR genome stanza (same HUMAN motifs as hs1):

hs1_chrR:
  indexdir: "/data/CCBR_Pipeliner/db/PipeDB/Indices/hs1_chrR/indexes"
  blacklistFa: "$PIPELINE_HOME/resources/blacklistFa/hs1_chrR.blacklist.fa.gz"
  chroms: "$PIPELINE_HOME/resources/chroms/hs1_chrR.chroms"
  tssBed: "$PIPELINE_HOME/resources/tssBed/hs1_chrR_tssbeds.tar.gz"
  homermotif: "$PIPELINE_HOME/resources/motif/HOCOMOCOv11_full_HUMAN_mono_homer_format_0.001.motif"
  mememotif: "$PIPELINE_HOME/resources/motif/HOCOMOCOv11_core_HUMAN_mono_meme_format.tar.gz"
  effectiveGenomeSize: 3000000000   # confirm actual value for chrR subset
  fripextra:
    dhsbed: "$PIPELINE_HOME/resources/frip/hs1_chrR.DHS.bed.gz"
    promoterbed: "$PIPELINE_HOME/resources/frip/hs1_chrR.promoters.bed.gz"
    enhancerbed: "$PIPELINE_HOME/resources/frip/hs1_chrR.enhancers.bed.gz"

5. bin/aspen

Add hs1_chrR to the SUPPORTED_GENOMES variable (line ~65):

SUPPORTED_GENOMES="hg19 hg38 mm10 mmul10 bosTau9 hs1 hs1_chrR"

6. config/config.yaml — user-facing comment

Uncomment/add hs1_chrR to the example genome comment block near line 30:

# genome: "hs1_chrR"

7. Documentation

  • Update docs/ (e.g., docs/introduction.md or docs/overview.md) to list hs1_chrR as a supported genome.
  • Add CHANGELOG.md entry under ## development version.

Notes

  • hs1_chrR inherits HUMAN motif files from hs1; no new motif files are needed.
  • effectiveGenomeSize should be confirmed once the chrR-subset FASTA is assembled (the hs1 value of 3,000,000,000 is a reasonable starting point but may differ for the canonical-chromosome subset).
  • frip and tssBed files for hs1_chrR reuse the hs1 originals (see reuse decision comment below).

Requested by

⚡ Generated using AI ⚡

Activity

  1. kopardev commented on Aug 3, 2026

    @kopardev
    ContributorAuthor

    Source genome clarification

    The specific genome version to be used for hs1_chrR is the Human_hs1-rDNA_genome_v1.0 from:

    George SS, Pimkin M, Paralkar VR.
    "Construction and validation of customized genomes for human and mouse ribosomal DNA mapping."
    JBC (2023). https://www.jbc.org/article/S0021-9258(23)01794-5/fulltext

    Repository: https://github.com/vikramparalkar/rDNA-Mapping-Genomes

    What this genome is:
    The hs1 (T2T-CHM13) assembly has been:

    1. Masked for rDNA-like regions throughout the canonical chromosomes.
    2. Augmented with an additional chrR containing a modified version of the NCBI Genbank KY962518.1 rDNA sequence — this is the rDNA reference unit used for quantifying ribosomal DNA accessibility in ATAC-seq.

    Relevant files from that repo:

    File Contents
    Human_hs1-rDNA_genome_v1.0.tar.gz (972 MB) FASTA + FAI: hs1-rDNA_v1.0.fa, hs1-rDNA_v1.0.fa.fai
    Human_hs1-rDNA_genome_v1.0_annotation.tar.gz RefSeq annotations + custom chrR annotations: hs1-rDNA_v1.0.bed
    Human_hs1-rDNA_genome_v1.0_control_tracks.tar.gz Validation BigWigs (CTCF, TBP, POLR1A, UBTF, EP300)

    Implications for resource file creation (Task 1–2 above):

    • The index directory FASTA should be derived from hs1-rDNA_v1.0.fa (not a plain hs1 FASTA).
    • resources/chroms/hs1_chrR.chroms should include all canonical hs1 chromosomes plus chrR.
    • The chrR-specific annotation BED (hs1-rDNA_v1.0.bed) should be used when building TSS/promoter/enhancer resource files; standard hs1 BED files should be filtered to exclude unmasked rDNA-like loci.
    • effectiveGenomeSize should be recalculated from the masked+chrR FASTA (likely slightly lower than the unmasked hs1 value of 3,000,000,000).
  2. self-assigned this
    on Aug 3, 2026
  3. kopardev commented on Aug 3, 2026

    @kopardev
    ContributorAuthor

    Resource file reuse decision

    The following hs1_chrR resource files will reuse the existing hs1 files directly (symlink or copy), since the standard chromosomes and annotation content are identical between hs1 and hs1_chrR — the only addition in hs1_chrR is the chrR chromosome:

    hs1_chrR file Reused from
    resources/frip/hs1_chrR.DHS.bed.gz resources/frip/hs1.DHS.bed.gz
    resources/frip/hs1_chrR.promoters.bed.gz resources/frip/hs1.promoters.bed.gz
    resources/frip/hs1_chrR.enhancers.bed.gz resources/frip/hs1.enhancers.bed.gz
    resources/tssBed/hs1_chrR_tssbeds.tar.gz resources/tssBed/hs1_tssbeds.tar.gz

    These files do not need to be regenerated from scratch. Tasks 2 checklist items for these four files can be resolved by copying/symlinking the hs1 originals and committing them under the hs1_chrR names via LFS.

    The chrR-specific annotation BED (hs1-rDNA_v1.0.bed from the Paralkar repo) only needs to be incorporated if rDNA-region FRiP or TSS enrichment metrics are desired for chrR specifically — that can be a follow-up enhancement.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions