sequencing_comparison

A. Compute coverage

Setup

1. Make sure that the latest versions of samtools, bcftools, and tabix are installed.
2. Download the latest Nextflow from https://www.nextflow.io.
3. Create python 3 virtual environment: virtualenv -p python3.6 venv.
4. Activate virtual environment: source venv/bin/activate.
5. Install python packagies: pip install -r requirements.txt.

Run Step 1: prepare intervals covering CDS regions (i.e. protein coding exons and some padding around them)

1. cd into intervals directry

2. download latest GENCODE GTF file:

   wget ftp://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_human/release_31/gencode.v31.annotation.gtf.gz

3. create intervals:

   python make_intervals.py -g gencode.v31.annotation.gtf.gz -o cds_intervals

   Intervals will be saved into cds_intervals directory. By default, chromosome names have chr prefix. If your reference genome doesn't have this prefix, then specify --no-chr-prefix option when running make_intervals.py.

Run Step 2: generate DP information

1. cd into coverage directory

2. edit nextflow.config file:

   bams_list_path -- points to the file were each line is a whitespace delimited tuple: sample name, absolute path to the corresponding BAM/CRAM file.

   cram -- true if working with CRAMs

   intervals -- points to *.list files with intervals generated in Step 1.

   reference_path -- points to FASTA file (*.fa) with genome reference. File must be indexed (ie. the corresponding *.fai should in the same directory)

   samtools -- path to samtools executable

   max_depth -- maximal depth per sample (see documentation for samtools mpileup for details)

   mpileup_subset -- points to mpileup_subset.py script (prefferable absolute path)

   bcftools -- path to bcftools executable

   tabix -- path to tabix executable

   Edit other options related to SLURM or local execution as needed.

3. Run nextlow run Coverage.nf. Preferrably run from tmux session. When crashed (e.g. SLURM node failure) use nextflow run Coverage.nf -resume.

4. The final BCF files with DP information for each CDS base-pair and each sample are located in results/merged folder.

B. Subset VCF and normalize variants

Setup

1. Make sure that the latest versions of bcftools and tabix are installed.
2. Download the latest Nextflow from https://www.nextflow.io.

Run

1. cd into subset_normalize directory

2. edit nextflow.config file:

   bams_list_path -- Same file as in coverage computation: points to the file were each line is a whitespace delimited tuple: sample name, absolute path to the corresponding BAM/CRAM file.

   vcfs -- path to VCF or BCF files with genotype information. Files must include both PASS and QC failed variants.

   reference_path -- points to FASTA file (*.fa) with genome reference. File must be indexed (ie. the corresponding *.fai should in the same directory)

   bcftools -- path to bcftools executable

   tabix -- path to tabix executable

   Edit other options related to SLURM or local execution as needed.

3. Run nextlow run Subset.nf. Preferrably run from tmux session. When crashed (e.g. SLURM node failure) use nextflow run Subset.nf -resume.

4. The final BCF files with normalized variants are located in results/ folder.

C. Annotate variants

Setup

1. Make sure that the latest version of bcftools and tabix are installed
2. Make sure that the VEP (Variant Effect Predictor) v96 or higher is installed.

Run

1. cd into annotate directory

2. edit nextflow.config file:

   vcfs -- path to the VCF/BCF files generate in the previous step

   vep -- path to VEP executable

   vep_flags -- set any additional VEP flags if needed

   bcftools -- path to VEP executable

   tabix -- path to tabix executable

   Edit other options related to SLURM or local execution as needed.

3. Run nextlow run Annotate.nf. Preferrably run from tmux session. When crashed (e.g. SLURM node failure) use nextflow run Annotate.nf -resume.

4. The final VCF files with annotated variants are located in results/vep folder.

D. Pairwise sample comparison (concordance)

1. cd into compare directory

2. edit nextlow.config file:

   pairs_list_path -- Mapping file with sample IDs (no header). Each line has two whitespace (or tab) delimited columns. First column stores sample ID in study 1, second column stores corresponding sample ID in study 2. See example_pairs.list.

   study1_files_list_path -- File with absolute paths to coverage, genotype, and annotation files generated in previous steps for study 1. The file has three whitespace (or tab) delimited columns (no header). The first column stores absolute path to coverage files from step A, the second column stores absolute path to the genotype files from step B, the third column stores absolute path to the annotation files from step C. Important: each row must store corresponding files for the same chromosome. See study1_files.list.

   study2_files_list_path -- File with the same structure as study1_files_list_path, but with absolute paths to coverage, genotype, and annotation files for study 2.

   compare -- absolute path to thecompare.py script

   Edit other options related to SLURM or local execution as needed.

3. Run nextlow run Compare.nf. Preferrably run from tmux session. When crashed (e.g. SLURM node failure) use nextflow run Compare.nf -resume.

4. The final gzip compressed summary files are located in results/ folder.

E. Depth histograms in CDS

1. cd into histograms directory

2. edit nextflow.config

   gencode_gtf_path -- path to GENCODE GTF file. Change only if other than v31 version needed.

   coverage_files_path -- path to VCF/BCF files generated in step A (i.e. in sequencing_comparison/coverage/results/merged directory).

   coverage_files_index_suffix -- change to tbi if VCF/BCF files in step A were indexed using TBI index (i.e. default tabix).

   histograms -- absolute path to thehistograms.py script.

   Edit other options related to SLURM or local execution as needed.

3. Run nextlow run Histograms.nf. Preferrably run from tmux session. When crashed (e.g. SLURM node failure) use nextflow run Histograms.nf -resume.

4. The final gzip compressed histogram files are located in results/ folder.