case.control.snake

#High Throughput Case Control 3D-Clump Pipeline
#Tychele Turner, Ph.D.
#Yilin Chen; Rachel Karchin, Ph.D., Lab Johns Hopkins University
#Jeffrey K. Ng; Tychele Turner, Ph.D., Lab Washington University in St. Louis

#Initial Commit: 2/13/23

import os

#configuration
if config == {}:
    configfile: "case.control.config.json"
PROTEIN_FILE = config["protein_file"]
MUTFILE = config["mutation_file"]
CONTROLS = config["control_file"]
PERMUTATIONS = config["permutations"]
PROTEIN_LENGTH = config["protein_length"]
PDB_DIR = config["structures_directory"]
#show proteins to run
PROTEINS = []
with open(PROTEIN_FILE) as f:
    PROTEINS = f.read().splitlines()
print(PROTEINS)

rule all:
	input: MUTFILE, expand("track/{protein}_finished.txt", protein = PROTEINS) 


rule case_control_clump:
        input: MUTFILE
	output: "track/{protein}_finished.txt"
        params: prefix="{protein}"
        shell: """
	     #path	     
	    export PATH=/opt/conda/envs/snakemake/bin:$PATH
        set +e
             mkdir -p input_genes
             mkdir -p control_genes
             mkdir -p clump_results
             mkdir -p track

             grep -w {params.prefix} {input} > input_genes/{params.prefix}
             grep -w {params.prefix} {CONTROLS} > control_genes/{params.prefix}
             python3 3D-CLUMP.py -s {PDB_DIR}  -f input_genes/{params.prefix} -p {PROTEIN_LENGTH} -c control_genes/{params.prefix} -m 5 -z {PERMUTATIONS} > clump_results/{params.prefix}
             touch {output}
	"""