Tools for introduction of machine learning to molecular dynamics simulations
Prerequisites:
MDAnalysis (pip3 install MDAnalysis)
mouse2 (pip3 install mouse2)
Usage:
Creating .pdb files for individual aggregates:
split_pdb.py [-h] [--r_neigh [R_neigh]] [--selection [QUERY]] INPUT [INPUT ...]
Determine clusters in the last time frame of a trajectory and write each one into a separate pdb file.
positional arguments: INPUT input file(s), the format will be guessed by MDAnalysis based on file extension
options: -h, --help show this help message and exit --r_neigh [R_neigh] neighbor cutoff --selection [QUERY] Consider only selected atoms, use MDAnalysis selection language
Example:
split_pdb.py somedata.data somedump.lammpsdump --selection "type 1"
PDB files named somedata-i.pdb will be generated, where i is the id of an aggregate, starting from 1.
Metadata manipulation:
rw_metadata.py [-h] [--phil [PHIL ...]] [--phob [PHOB ...]] FILE ACTION [type]
Add structure metadata to pdb
positional arguments: FILE pdb file ACTION write/read/clear type Structure type
options: -h, --help show this help message and exit --phil [PHIL ...] Solvophilic bead types --phob [PHOB ...] Solvophobic bead types
Example:
rw_metadata.py somedata-1.pdb write TORUS --phil 1 --phob 2
rw_metadata.py somedata-1.pdb read
rw_metadata.py somedata-1.pdb clear