Skip to content

Latest commit

 

History

History
29 lines (24 loc) · 1.39 KB

workflows.md

File metadata and controls

29 lines (24 loc) · 1.39 KB

Computational Workflows

The workflows consist of the following steps:

  1. Inputs & Configuration:
  2. MD simulations
  3. Conformer generators
  4. Per compound analysis
  5. Per method analysis

Inputs

To simulate a compound, a new row needs to be added to the samples.tsv file. The row needs to contain the following information:

  • compound-id: unique identifier for the compound
  • method: MD-method used for the simulation
  • solvent: solvent to use for the simulation. Options: aqueous, DMSO, DMSO_GAFF_BCC, DMSO_GAFF_RESP, Chloroform
  • simtime: simulation time in ns
  • dt: timestep in ps
  • repeats: number of simulation repeats
  • other_param: This is a dummy parameter, changing it from 0.2 leads to creation of a different hash, so that the simulation is run again without overwriting the previous results.
  • igamd: boosting parameter for GaMD. 1: dihedral angles boost only, 2: total energy boost only, 3: both dihedral and total energy boost

The workflow automatically expands rows found in samples.tsv and produces a unique hash for every MD simulation (see the below figure). The hash is used to name the output files. Hashes can be found together with the parameters that were set in the samples_old.tsv file.

Expanding rows in samples.tsv

Configuration

Via the configuration file snakemake-config.yaml a number of parameters can be set.