Skip to content

README.md

Latest commit

 

History

History
52 lines (39 loc) · 2.15 KB

README.md

File metadata and controls

52 lines (39 loc) · 2.15 KB

Example Workflow: Submitting fpocketR Batch Jobs with Bash

This directory demonstrates how to use a simple bash script to queue and run multiple fpocketR jobs back-to-back. This approach is useful for users who want to submit complex fpocketR commands in batches, without the need for workflow managers like Snakemake.

Purpose

Show how to automate the submission of many fpocketR jobs using a bash script. This method is flexible for custom arguments, but runs jobs sequentially (slower than parallelized workflows).

Prerequisites

  • fpocketR installed (see main package instructions)
  • The fpocketR conda environment needs to be installed on your system
  • Python (recommended: use the version provided in the fpocketR environment)

File Structure

  • fpocketR_batch_file.txt: Text file listing fpocketR command-line arguments (one job per line)
  • fpocketR_batch_submitter.sh: Bash script to submit each job in sequence
  • Output directories: Created by fpocketR for each job

Workflow Steps

1. Prepare Your Batch File

List each fpocketR job as a line in fpocketR_batch_file.txt. Example:

-pdb 3e5c.pdb -o SAM-III_RS -y
-pdb 2l1v.pdb -ss 2l1v.nsd -o preQ1_RS -y
-pdb 2l1v.pdb -s 0 -ss 2l1v.nsd -o preQ1_RS_multistate -y
-pdb 8f4o_apo.pdb -al 2gdi_holo.pdb -nt 19,20,42,43 -o TPP_RS_apo_holo -y
-pdb 2gdi.pdb --chain Y --ligand TPP --dpi 10 --out TPP_RS -y

2. Run the Batch Submitter Script

Use the provided bash script to run all jobs in sequence:

bash fpocketR_batch_submitter.sh

This will:

  • Read each line from fpocketR_batch_file.txt
  • Run python -m fpocketR with the specified arguments
  • Print each command before running it

3. Review Results

  • Output files for each job are created in the specified output directories
  • Check the terminal output for job progress and errors

Notes

  • This method is ideal for custom or complex fpocketR arguments
  • Jobs run one after another (not in parallel)
  • For parallel performance, consider using Snakemake

For more details, see the fpocketR documentation.