00.21-E-Peve-BS-genome
================
Sam White
2025-01-02

- <a href="#1-background" id="toc-1-background">1 Background</a>
- <a href="#2-inputs" id="toc-2-inputs">2 Inputs</a>
- <a href="#3-outputs" id="toc-3-outputs">3 Outputs</a>
- <a href="#4-create-a-bash-variables-file"
  id="toc-4-create-a-bash-variables-file">4 Create a Bash variables
  file</a>
- <a href="#5-bisfulite-conversion" id="toc-5-bisfulite-conversion">5
  Bisfulite conversion</a>
  - <a href="#51-inpect-bs-output" id="toc-51-inpect-bs-output">5.1 Inpect
    BS output</a>
  - <a href="#52-compress-output-folder"
    id="toc-52-compress-output-folder">5.2 Compress output folder</a>
  - <a href="#53-create-md5sum" id="toc-53-create-md5sum">5.3 Create
    MD5sum</a>
- <a href="#6-references" id="toc-6-references">6 REFERENCES</a>

# 1 Background

This Rmd file will create a bisulfite-converted genome by, and for,
Bismark (Krueger and Andrews 2011) using the `Porites_evermanni_v1.fa`
file. The genome FastA was taken from the [Genoscop corals
webpage](https://www.genoscope.cns.fr/corals/genomes.html).

Due to large sizes of output files, the files cannot be sync’d to
GitHub. As such, the output directories will be gzipped and available
here:

- <https://gannet.fish.washington.edu/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome.tar.gz>
  (1.5GB)

- <https://gannet.fish.washington.edu/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome.tar.gz.md5>

  - MD5: `5a0d4f699d7d46eb9f996e677841582a`

# 2 Inputs

- Directory containing a FastA file with the file extension: .fa or
  .fasta (also ending in .gz).

# 3 Outputs

- CT Conversion

  - Bowtie2 index files.
  - CT conversion FastA

- GA conversion

  - Bowtie2 index files.
  - GA conversion FastA.

# 4 Create a Bash variables file

This allows usage of Bash variables across R Markdown chunks.

``` bash
{
echo "#### Assign Variables ####"
echo ""

echo "# Data directories"
echo 'export timeseries_dir=/home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular'
echo 'export output_dir_top=${timeseries_dir}/E-Peve/data'
echo 'export genome_dir=${timeseries_dir}/E-Peve/data'
echo ""

echo "# Paths to programs"
echo 'export programs_dir="/home/shared"'
echo 'export bismark_dir="${programs_dir}/Bismark-0.24.0"'
echo 'export bowtie2_dir="${programs_dir}/bowtie2-2.4.4-linux-x86_64"'
echo ""

echo "# Set number of CPUs to use"
echo 'export threads=20'
echo ""

echo "# Print formatting"
echo 'export line="--------------------------------------------------------"'
echo ""
} > .bashvars

cat .bashvars
```

    #### Assign Variables ####

    # Data directories
    export timeseries_dir=/home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular
    export output_dir_top=${timeseries_dir}/E-Peve/data
    export genome_dir=${timeseries_dir}/E-Peve/data

    # Paths to programs
    export programs_dir="/home/shared"
    export bismark_dir="${programs_dir}/Bismark-0.24.0"
    export bowtie2_dir="${programs_dir}/bowtie2-2.4.4-linux-x86_64"

    # Set number of CPUs to use
    export threads=20

    # Print formatting
    export line="--------------------------------------------------------"

# 5 Bisfulite conversion

``` bash
# Load bash variables into memory
source .bashvars

${bismark_dir}/bismark_genome_preparation \
${genome_dir} \
--parallel ${threads} \
--bowtie2 \
--path_to_aligner ${bowtie2_dir} \
1> ${genome_dir}/Peve-bs-genome.stderr
```

    Using 20 threads for the top and bottom strand indexing processes each, so using 40 cores in total
    Writing bisulfite genomes out into a single MFA (multi FastA) file

    Bisulfite Genome Indexer version v0.24.0 (last modified: 19 May 2022)

    Step I - Prepare genome folders - completed



    Step II - Genome bisulfite conversions - completed


    Bismark Genome Preparation - Step III: Launching the Bowtie 2 indexer
    Building a SMALL index
    Building a SMALL index

    =========================================

    Parallel genome indexing complete. Enjoy!

## 5.1 Inpect BS output

``` bash
# Load bash variables into memory
source .bashvars

tree -h ${genome_dir}/Bisulfite_Genome
```

    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome
    ├── [4.0K]  CT_conversion
    │   ├── [184M]  BS_CT.1.bt2
    │   ├── [134M]  BS_CT.2.bt2
    │   ├── [442K]  BS_CT.3.bt2
    │   ├── [134M]  BS_CT.4.bt2
    │   ├── [184M]  BS_CT.rev.1.bt2
    │   ├── [134M]  BS_CT.rev.2.bt2
    │   └── [586M]  genome_mfa.CT_conversion.fa
    └── [4.0K]  GA_conversion
        ├── [184M]  BS_GA.1.bt2
        ├── [134M]  BS_GA.2.bt2
        ├── [442K]  BS_GA.3.bt2
        ├── [134M]  BS_GA.4.bt2
        ├── [184M]  BS_GA.rev.1.bt2
        ├── [134M]  BS_GA.rev.2.bt2
        └── [586M]  genome_mfa.GA_conversion.fa

    2 directories, 14 files

## 5.2 Compress output folder

``` bash
source .bashvars

tar -czvf ${genome_dir}/Bisulfite_Genome.tar.gz ${genome_dir}/Bisulfite_Genome
```

    tar: Removing leading `/' from member names
    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome/
    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome/CT_conversion/
    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome/CT_conversion/BS_CT.4.bt2
    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome/CT_conversion/BS_CT.rev.1.bt2
    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome/CT_conversion/genome_mfa.CT_conversion.fa
    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome/CT_conversion/BS_CT.1.bt2
    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome/CT_conversion/BS_CT.2.bt2
    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome/CT_conversion/BS_CT.3.bt2
    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome/CT_conversion/BS_CT.rev.2.bt2
    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome/GA_conversion/
    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome/GA_conversion/BS_GA.rev.1.bt2
    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome/GA_conversion/BS_GA.2.bt2
    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome/GA_conversion/BS_GA.4.bt2
    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome/GA_conversion/genome_mfa.GA_conversion.fa
    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome/GA_conversion/BS_GA.rev.2.bt2
    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome/GA_conversion/BS_GA.1.bt2
    /home/shared/8TB_HDD_01/sam/gitrepos/urol-e5/timeseries_molecular/E-Peve/data/Bisulfite_Genome/GA_conversion/BS_GA.3.bt2

## 5.3 Create MD5sum

``` bash
source .bashvars

cd ${genome_dir}

md5sum Bisulfite_Genome.tar.gz | tee Bisulfite_Genome.tar.gz.md5
```

    1f65833895ba3c8d50fe27bb8ad5303e  Bisulfite_Genome.tar.gz

# 6 REFERENCES

<div id="refs" class="references csl-bib-body hanging-indent">

<div id="ref-krueger2011" class="csl-entry">

Krueger, Felix, and Simon R. Andrews. 2011. “Bismark: A Flexible Aligner
and Methylation Caller for Bisulfite-Seq Applications.” *Bioinformatics*
27 (11): 1571–72. <https://doi.org/10.1093/bioinformatics/btr167>.

</div>

</div>