A python-based miRNA sequencing pipeline for isomiR quantification and analysis

• Make sure that you have Python 3.4+ installed (type python --version in the console)
• Make sure you have the latest version of pip: pip3 install -U pip
• Make sure you have Miniconda installed

1. Download repository: git clone https://github.com/Gu-Lab-RBL-NCI/QuagmiR
2. Go into local quagmir folder: cd quagmir
3. Install Python dependencies: conda env create -f environment.yml

1. Add your .fastq samples into the data folder (a sample has been provided for testing):

  ├── LICENSE
  ├── README.md
  ├── config.yaml
  ├── Snakefile
  ├── environment.yml
  ├── data/
  │   ├── sample.fastq_ready
  │   └── YOUR_FILE_HERE.fastq_ready
  |   ├── collapsed/
  ├── motif-consensus.fa
  └── results/
  │   └── tabular/
  

2. Edit the motif-consensus.fa file to insert your miRNA information with the following format:

>miRNA_name miRNA_motif
miRNA_consensus_sequence

>passenger-shRNA ACACCCTGGCCGGGT
CCGACACCCTGGCCGGGTTGT

3. Edit the config.yaml file to change configuration options if needed (default values fine in most use cases):

More information about configuration parameters.

# FILTERING
min_ratio: -1 # to be set to -1 to switch off min_ratio filtering
min_read: -1 # to be set to -1 to switch off min_read filtering
edit_distance_3p: -1 # to be set to -1 to switch off 3p filtering
edit_distance_5p: -1 # to be set to -1 to switch off 5p filtering

# DISPLAY
display_summary: True
display_sequence_info: True
display_nucleotide_dist: True
display_group_output: True
display_expression: True
display_distance_metric: True

# AMBIGUOUS LETTERS
ambiguous_letters: True

# DESTRUCTIVE MOTIF PULL
destructive_motif_pull: False

# INPUT
motif_consensus_file: 'motif_consensus_file.fa'

# OUTPUT
group_output_name: 'cohort1'

# DISTANCE METRIC
deletion_score: 1
insertion_score: 1
substitution_AG: 1 # transition
substitution_GA: 1 # transition
substitution_CT: 1 # transition
substitution_TC: 1 # transition
substitution_AT: 1 # transversion
substitution_TA: 1 # transversion
substitution_AC: 1 # transversion
substitution_CA: 1 # transversion
substitution_GC: 1 # transversion
substitution_CG: 1 # transversion
substitution_GT: 1 # transversion
substitution_TG: 1 # transversion

4. Run pipeline: bash run.sh or activate conda env with source activate quagmir and run snakemake

Run the commands git reset --hard and git pull.

Collapsing the sample fastq files has the longest runtime. But once the samples are collapsed, rerunning the pipeline will be much faster as the files will not be recollapsed.