assembly_qc

assembly_qc is a Snakemake workflow for assembly quality control. The workflow starts from haplotype-resolved assembly FASTA files, performs contamination/adaptor/rDNA cleaning, then generates assembly statistics, reference-alignment summaries, QV/completeness metrics, BUSCO/compleasm summaries, and human-specific visualization outputs.

Main outputs

For each sample and available haplotype, the workflow can generate:

• cleaned FASTA files after short-contig filtering, FCS-GX/FCS-adaptor screening, optional mitochondrial extraction, rDNA splitting, and final FASTA renaming
• scaffold/contig assembly statistics, including N50-style summaries and telomere annotations
• assembly-to-reference PAF, SAF, chromosome coverage summaries, and chain files
• Merqury QV and completeness outputs from Illumina reads
• optional trio Merqury hap-mer plots when parental IDs are provided
• compleasm/BUSCO gene completeness summaries
• contig length plots, ideograms, ploidy plots, and assembly-eval config files
• human-only ModDotPlot summaries for acrocentric regions
• optional SMaHT DSA metadata tables when SMAHT_METADATA: True

Repository layout

.
├── Snakefile
├── config/
│   ├── config_asm_qc.yaml
│   └── manifest_asm_qc.tab
├── rules/
│   ├── merqury.smk
│   ├── saffire.smk
│   ├── compleasm.smk
│   ├── fasta_stats.smk
│   ├── moddotplot.smk
│   └── plots.smk
├── scripts/
├── runcluster
└── runlocal

Configuration

The default config is:

MANIFEST: config/manifest_asm_qc.tab

REF:
  CHM13:
    PATH: /net/eichler/vol28/eee_shared/assemblies/CHM13/T2T/v2.0/T2T-CHM13v2.fasta
    CHROMS: /net/eichler/vol28/eee_shared/assemblies/CHM13/T2T/v2.0/genome.txt
    CYTO: /net/eichler/vol28/eee_shared/assemblies/CHM13/T2T/v2.0/anno/cyto.bed
  GRCh38:
    PATH: /net/eichler/vol28/eee_shared/assemblies/hg38/no_alt/hg38.no_alt.fa
    CHROMS: /net/eichler/vol28/eee_shared/assemblies/hg38/no_alt/genome.txt
    CYTO: /net/eichler/vol28/eee_shared/assemblies/hg38/no_alt/anno/cyto.bed

DB_PATH: /net/eichler/vol28/eee_shared/buscodb/
LINEAGE: primates
MODE: busco

TAXID: 9606
INCLUDE_MITO: True
HPRC_NAMING: True

Key config fields

Field	Description
MANIFEST	Tab-delimited manifest describing assemblies, read FOFNs, and optional trio information.
REF	Reference genomes to use for minimap2/Saffire outputs, ideograms, ploidy plots, and chain files. Each reference requires PATH and CHROMS; CYTO is optional.
DB_PATH	BUSCO/compleasm database path.
LINEAGE	compleasm lineage, such as primates.
MODE	compleasm mode, such as busco.
TAXID	Expected organism taxid for FCS-GX screening. Human-specific downstream outputs are enabled for 9606.
INCLUDE_MITO	Whether mitochondrial FASTA extraction should be included when supported by the workflow settings.
HPRC_NAMING	Whether final FASTA names should follow HPRC-style naming logic.

Manifest

The manifest is tab-delimited. Current expected columns are:

SAMPLE  NAME  H1  H2  UNASSIGNED  ILLUMINA  FOFN  TRIO  MO_ID  FA_ID

Example:

SAMPLE_verkko  SAMPLE  test_data/asm/verkko.hap1.fasta  test_data/asm/verkko.hap2.fasta  test_data/asm/verkko.unassigned.fasta  SAMPLE  test_data/fofn/reads.fofn  NO  NA  NA

Manifest columns

Column	Description
SAMPLE	Workflow sample ID. This is used in results/{SAMPLE}.
NAME	Display or output naming value used by the workflow.
H1	Haplotype 1 assembly FASTA. Required for a sample to be included in the main workflow.
H2	Haplotype 2 assembly FASTA. Optional.
UNASSIGNED	Unassigned assembly FASTA. Optional.
ILLUMINA	Illumina read sample ID used to locate/read the FOFN for Merqury. Multiple assemblies can share the same Illumina ID.
FOFN	File of file names for Illumina FASTQ inputs used by Merqury/meryl. One FASTQ path per line.
TRIO	YES or NO; enables trio Merqury outputs when parental IDs are available.
MO_ID	Maternal Illumina sample ID for trio Merqury. Use NA if not available.
FA_ID	Paternal Illumina sample ID for trio Merqury. Use NA if not available.

Running the workflow

Run a dry-run first:

./runcluster 30 -np

Run on the cluster:

./runcluster 30

Run locally:

./runlocal 30

runcluster and runlocal both resolve the repository path automatically, load miniconda/4.12.0, create log/, and pass the remaining arguments to Snakemake. runcluster submits jobs through DRMAA/SGE with mfree, h_rt, threads, and -V -cwd -j y settings. Both wrappers use Singularity/Apptainer and bind /net.

To use a different config file:

./runcluster 30 --configfile path/to/config.yaml

Partial workflow targets

You can run selected output groups instead of the full workflow:

./runcluster 30 get_cleaned_fasta_only
./runcluster 30 get_stats_only
./runcluster 30 get_saf_only
./runcluster 30 get_qv_only
./runcluster 30 get_busco_only
./runcluster 30 get_plots_only
./runcluster 30 get_moddotplots_only

Target	Outputs
get_cleaned_fasta_only	Final cleaned FASTA files under results/{sample}/contamination_screening/outputs/final_fasta/.
get_stats_only	Per-haplotype FASTA statistics under results/{sample}/stats/outputs/summary/.
get_saf_only	Saffire SAF files, chromosome coverage summaries, trimmed PAFs, and human chain files.
get_qv_only	Merqury QV and completeness outputs. Trio plots are included when TRIO=YES.
get_busco_only	compleasm/BUSCO summary TSV.
get_plots_only	Contig length plots, ideograms, ploidy plots, and assembly-eval config files.
get_moddotplots_only	Human acrocentric ModDotPlot summary and contig statistics.

Workflow overview

The full all target creates:

results/{sample}/complete_flag/all_done

for every sample with a valid H1 assembly in the manifest.

Internally, the workflow includes these rule modules:

1. fcs_gx.smk
   Links/indexes raw FASTA input, removes very short contigs, runs FCS-GX/adaptor screening, detects mitochondrial and rDNA sequence, splits rDNA, and writes final cleaned FASTA outputs.

2. merqury.smk
   Builds meryl databases from Illumina FASTQ FOFNs, computes Merqury QV/completeness, and optionally generates trio hap-mer outputs.

3. saffire.smk
   Prepares reference FASTA resources, aligns cleaned assemblies to each configured reference with minimap2, trims/filters PAFs, generates SAF/chromosome coverage summaries, and creates chain files for human assemblies.

4. compleasm.smk
   Runs compleasm and summarizes BUSCO-style completeness metrics.

5. fasta_stats.smk
   Splits scaffolds into contigs, calculates telomere annotations, computes per-haplotype and full-genome assembly statistics, and summarizes stats tables.

6. moddotplot.smk
   For human assemblies, identifies acrocentric contigs from CHM13 alignments and generates ModDotPlot-related summaries/plots.

7. plots.smk
   Generates contig length plots, ideogram plots, ploidy plots, and assembly-eval configuration files.

Important output locations

results/{sample}/
├── complete_flag/
│   └── all_done
├── contamination_screening/
│   └── outputs/
│       ├── final_fasta/{sample}_{hap}.fasta
│       ├── contig_fasta/{sample}_{hap}.fasta
│       ├── mito_fasta/{hap}-mt.fasta
│       └── rdna_fasta/{hap}-rdna.fasta
├── stats/
│   └── outputs/
│       └── summary/{hap}.summary.stats
├── saffire/
│   └── outputs/
│       ├── trimmed_pafs/{ref}/{hap}.minimap2.trimmed.paf
│       ├── safs/{ref}/{hap}.minimap2.saf
│       └── chrom_cov/{ref}/{hap}.minimap2.chrom_cov.tsv
├── chain_files/
│   └── outputs/
│       ├── {sample}_{hap}_To_{ref}.chain
│       └── {sample}_{hap}_To_{ref}.invert.chain
├── merqury/
│   └── outputs/
│       ├── {sample}.qv
│       └── {sample}_{hap}.qv
├── compleasm/
│   └── outputs/
│       └── summary/{sample}.summary.tsv
├── plots/
│   └── outputs/
│       ├── contig_length/{hap}.scaffold.scatter_logged.png
│       ├── ideo/{ref}/pdf/{sample}.minimap2.ideoplot.pdf
│       ├── ideo/{ref}/pdf/{sample}.minimap2.ideoplot_wide.pdf
│       └── ploidy/{ref}/pdf/{sample}.minimap2.ploidy.pdf
└── moddotplot/
    └── outputs/
        ├── summary/{hap}.generated_acros.tsv
        └── contig_stats/{hap}.CHM13_lifted_contigs.tsv

Shared resources are written under:

resources/
├── reference/{ref}/
│   ├── genome.fa
│   ├── genome.fa.fai
│   └── genome_index.txt
├── meryl/{ILLUMINA}/
└── acro_target_beds/

Notes

• Human-specific outputs, including ModDotPlot and chain files, depend on TAXID: 9606.
• H1 is required for a sample to be selected by the default workflow.
• H2 and UNASSIGNED are optional; the workflow runs only for haplotypes present in the manifest.
• FOFN is used for Merqury/meryl input and should list Illumina FASTQ files one per line.
• The workflow uses Singularity/Apptainer containers through Snakemake’s --use-singularity option.