Deconvolution

This page provides an overview of the deconvolution process in BLADE and instructions on how to apply it to your data using the provided Snakemake pipeline.

---

Steps for Deconvolution Using Snakemake

1. Prepare Your Environment

1. Ensure that all dependencies are installed, including Snakemake and the required Conda environments.

   • Install Snakemake:

     conda install -c bioconda -c conda-forge snakemake
   • Verify Conda environments:
     • Ensure you have an environment file (oncoBLADE.yaml) for running BLADE.
     • Create the environment if it doesn't already exist:

       conda env create -f envs/oncoBLADE.yaml

2. Prepare the directory structure:

   • Ensure the Bulk_data directory contains your bulk RNA-seq data in the file Transcriptome_matrix.txt.
   • Ensure the SignaturePipeline/output directory contains the signature matrix file Corrected_Signature_matrix.pickle.

3. Update the config.yaml file:

   • Edit the config.yaml file to specify the data directory and other parameters:

     all:
       data_dir: '/net/beegfs/scratch/abose/BLADE'
     
     BLADE:
       Parameters:
         Alpha: 1
         Alpha0: 1000
         Kappa0: 1
         SY: 1
         Nrep: 1  
         Nrepfinal: 1
         Njob: 10        
         std_const: 0.001

---

2. Run the Workflow

Use Snakemake to execute the workflow:

snakemake -s Snakefile \
    --cores 10 \
    --use-conda