Jeongho Chae
Benjamin McMichael
Date: 2025-01-24

This analysis utilises a snakemake pipeline to process ATAC-seq data. Once the pipeline has been cloned to the analysis directory (preferably in scratch space) using the command:

git clone https://sc.unc.edu/dept-fureylab/wasp_chromatin.git

There's no only prerequisite for running and the command for executing the WASP_chromatin pipeline is:

sbatch Snakemake_SLURMsubmission.sbatch

• Updated module versions to reflect newer versions of tools
• Organized the temp directory and final results directory for clarify
• Implement the moveout function

Used Previous versions for comparison with past WASP results.
Pervious versions should be used together rather than mixing them with current versions.

• WASP
  • Previous WASP version : 2019.12
  • Current WASP version : 2023.02
• python
  • Previous python version : 3.6.6
  • Current python version : 3.9.6
• bowtie2
  • Previous bowtie2 version : 2.4.1
  • Current bowtie2 version : 2.4.5
• samtools
  • Previous samtools version : 1.12
  • Current samtools version : 1.21

• Updated Perl scripts and created a snakemake pipeline
• Scripts used previously in the analysis of a subset of ATAC data can be found here:

/proj/fureylab/projects/CD_allelic_imbalance/wasp/wasp_scripts

• WASP uses genotype data to infer snp's and allelic usage
• Genotype data for use with WASP can be found here:

/proj/fureylab/data/Genotypes/human/imputed_vcfs_hg38

Implemented WASP so that the rmdup_pe rule is not executed redundantly, assuming that the reads for WASP have already had duplicates removed. However, if removeDupReads is set to TRUE in project_config.yaml, the remove duplicates rule will be executed.

1. Rule all

• Defines the final expected output files.

2. Rule find_intersecting_snps

• Identifies sequencing reads that overlap known SNPs.
• Generates three sets of reads:
  • Reads that require remapping (alternative alleles substituted) -> FASTQ file
  • Reads that require remapping (reference alleles retained) -> BAM file
  • Reads that do not require remapping -> BAM file

3. Rule remap_bowtiew2

• Re-aligns the remapping reads to the reference genome using Bowtie2.
• Produces a BAM file with newly mapped reads.

4. Rule sort_index_remapBam

• Sorts and indexes the remapped BAM file for efficient processing.

5. Rule filter_remapped_reads

• Compares remapped reads with their original versions.
• Discard reads that do not map back to their original locations, reducing allele-specific mapping bias.

6. Rule merge_bams

• Merges the filtered remapped BAM file with the original non-remapped BAM file generated from Rule 2.

For Rule 7 and 8, the rules will be executed only when removeDupReads is set to TRUE in project_config.yaml

7. Rule sort_index_mergeBam

• Sorts and indexes the merged BAM file for efficient processing.

8. Rule rmdup_pe

• Removes PCR duplicates from paired-end reads.
• Keeps only unique reads to prevent bias in allele-spcific analysis.

9. Rule sort_index_rmdup_pe

• Sorts and indexes the final BAM file for downstream analysis.

Moves final results files for locally run samples to permenant space. set moveOutFiles to TRUE in project_config.yaml after running the pipeline, checking that everything had run correctly, and then rerun the pipeline using the same submission statement.