/pipeline_DTU

Pipeline to determine differential transcript usage (DTU) from RNA-sequencing data and detects flips in isoform expression between experimental conditions.

Primary LanguagePython

Pipeline DTU

Usage

To use pipeline_DTU to investigate DTU of utrons (DTUtrons) use one of the following commands:

"python <pipeline_location>/pipeline_DTU.py make DTUtrons -v5" <-- local interactive
"submit_pipeline <pipeline_location>/pipeline_DTU.py make DTUtrons -v5" <-- new session

To use pipeline_DTU to investigate DTU for whole transcripts use one of the following commands:

"python <pipeline_location>/pipeline_DTU.py make DTU -v5" <-- local session
"submit_pipeline <pipeline_location>/pipeline_DTU.py make DAS -v5" <-- new session

Configuration

The pipeline requires a configured 'pipeline.yml' file. An example can be found at:

"<pipeline_dir>/pipeline_DTU/pipeline.yml"

Requirements

The pipeline requires the results from:

  • pipeline_fastqc
  • pipeline_mapping

How to use:

  • Input files in '.bam' format should be placed into an "input_assemble.dir"
  • Complete/modify the pipeline.yml
    • Importantly, information in the "dtu" section will need to be modified on a use-by-use basis
    • Including details on:
      • Whether data points are paired / unpaired
      • Whether DTU is based upon effect of one variable (univariate analysis) or two (covariate analysis)
      • The design formula that the linear model will be based around
      • Significance thresholds (adjusted p-values and log2fold change thresholds)
  • Design matrix must be present in the working directory
    • This allows mutliple comparisons to be made in one go
    • An example is show in "<pipeline_dir>/pipeline_DTU/design.tsv"