Bisulfite Sequencing Pipeline

A Nextflow pipeline to align and quantify Methylation (Bisulfite) sequencing data.

The pipeline was created to run on the ETH Euler cluster and it relies on the server's genome files. Thus, the pipeline needs to be adapted before running it in a different HPC cluster.

Pipeline steps

1. FastQC
2. FastQ Screen
3. Trim Galore
4. FastQC
5. Bismark
6. Bismark filter non-conversion [Optional]
7. Bismark deduplication
8. Bismark methylation extractor
9. coverage2cytosine [Optional]
10. Bismark2report
11. Bismark2summary
12. MultiQC

Required parameters

Path to the folder where the FASTQ files are located.

--input /cluster/work/nme/data/josousa/project/fastq/*fastq.gz

Output directory where the files will be saved.

--outdir /cluster/work/nme/data/josousa/project

Genomes

• Reference genome used for alignment.

  --genome

  Available genomes:

      Mus_musculus_GRCm39 # Default
      Mus_musculus_GRCm38_p6
      Homo_sapiens_GRCh38_p14
      Rattus_norvegicus_mRatBN7_2
      Bos_taurus_ARS-UCD1_2
      Bos_taurus_ARS-UCD1_3
      Caenorhabditis_elegans_WBcel235
      Callithrix_jacchus_mCalJac1_pat_X
      Capra_hircus_ARS1
      Capreolus_capreolus_GCA_951849835_1
      Escherichia_coli_ASM160652v1
      Macaca_fascicularis_Macaca_fascicularis_6_0
      Macaca_mulatta_Mmul_10
      Monodelphis_domestica_ASM229v1
      Pan_troglodytes_Pan_tro_3_0
      Saccharomyces_cerevisiae_R64-1-1
      Sus_scrofa_Sscrofa11_1

• Option to use a custom genome for alignment by providing an absolute path to a custom genome file.

  --custom_genome_file '/cluster/work/nme/data/josousa/project/genome/GRCm39.genome'

  Example of a genome file:

  name           GRCm39
  species        Mouse
  bismark        /cluster/work/nme/genomes/Mus_musculus/Ensembl/GRCm39/Sequence/BismarkIndex/          

FastQ Screen optional parameters

• Option to provide a custom FastQ Screen config file.

  --fastq_screen_conf '/cluster/work/nme/software/config/fastq_screen.conf' # Default

Bismark optional parameters

• Option to set the alignment mode to local.

  --local

  In this mode, it is not required that the entire read aligns from one end to the other. Rather, some characters may be omitted (“soft-clipped”) from the ends in order to achieve the greatest possible alignment score.

• Option to write all reads that could not be aligned to a file in the output directory.

  --unmapped

• Option to write all reads which produce more than one valid alignment with the same number of lowest mismatches or other reads that fail to align uniquely to a file in the output directory.

  --ambiguous

Skipping and adding options

• Option to skip FastQC, TrimGalore, and FastQ Screen. The first step of the pipeline will be the Bismark alignment. --skip_qc

• Option to skip FastQ Screen. --skip_fastq_screen

• Option to skip Bismark deduplication. --skip_deduplication

• Option to add Bismark filter non-conversion before deduplication (if selected) and before Bismark methylation extractor. --add_filter_non_conversion

Extra arguments

• Option to add extra arguments to FastQC. --fastqc_args

• Option to add extra arguments to FastQ Screen. --fastq_screen_args

• Option to add extra arguments to Trim Galore. --trim_galore_args

• Option to add extra arguments to Bismark. --bismark_arg

• Option to add extra arguments to Bismark filter non-conversion. --filter_non_conversion_args

• Option to add extra arguments to Bismark deduplication. --deduplicate_bismark_args

• Option to add extra arguments to Bismark methylation extractor. --bismark_methylation_extractor_args

• Option to add extra arguments to Bismark coverage2cytosine. --coverage2cytosine_args

• Option to add extra arguments to Bismark2summary. --bismark2summary_args

• Option to add extra arguments to Bismark2report. --bismark2report_args

• Option to add extra arguments to MultiQC. --multiqc_args

Acknowledgements

This pipeline was adapted from the Nextflow pipelines created by the Babraham Institute Bioinformatics Group and from the nf-core pipelines. We thank all the contributors for both projects. We also thank the Nextflow community and the nf-core community for all the help and support.