Create exome filter BED files and generation script

[hyunhwan-bcm]

Description

We need to create two versions of exome filter BED files (exon-only and gene-only), add them to our data dependencies, and include a script to generate these files in the util directory. This will provide more flexibility in filtering options for our pipeline users.

Proposed Changes

1. Create two BED files:
   • Exon-only filter
   • Gene-only filter (including both intron and exon)
2. Add these files to our data dependencies
3. Create a script to generate these files and add it to the util directory
4. Update documentation to reflect these new filtering options

Implementation Details

1. Create BED files

We need to create two BED files for each reference genome (hg19 and hg38):

• exon_only_hg19.bed and exon_only_hg38.bed
• gene_only_hg19.bed and gene_only_hg38.bed

These files should follow the standard BED format:

chromosome  start  end  [name]  [score]  [strand]

2. Add files to data dependencies

Add the following files to the data dependencies:

data_dependencies/
├── ref_exonic_filter_bed/ # the directory might not be the right one, please check
│   ├── hg19/
│   │   ├── exon_only.bed
│   │   └── gene_only.bed
│   └── hg38/
│       ├── exon_only.bed
│       └── gene_only.bed

Update the AWS S3 bucket with these new files:

aws s3 cp exon_only.bed s3://aim-data-dependencies-public/ref_exonic_filter_bed/hg19/
aws s3 cp gene_only.bed s3://aim-data-dependencies-public/ref_exonic_filter_bed/hg19/
aws s3 cp exon_only.bed s3://aim-data-dependencies-public/ref_exonic_filter_bed/hg38/
aws s3 cp gene_only.bed s3://aim-data-dependencies-public/ref_exonic_filter_bed/hg38/

3. Create generation script

Create a script named generate_exome_filters.py (or generate_exome_filters.R) in the util directory with the following structure:

4. Update documentation

Update the README and relevant documentation to include information about the new filtering options and how to use them in the pipeline.

Tasks

• [ ] Generate the four BED files (2 versions for each genome assembly)
• [ ] Add the generated BED files to the data dependencies
• [ ] Update the AWS S3 bucket with the new files
• [ ] Update the README and --help and documentation to reflect the new filtering options

Additional Notes

• Ensure that the generated BED files are sorted and non-overlapping
• Consider compressing the BED files to save space (e.g., using bgzip)
• Consider adding an option in the pipeline to choose between exon-only, gene-only, or no filtering

[jylee-bcm]

Resolved with the PR #55