LME: Layered Molecule Editor

LME is a molecule model editor designed with layers.

Plan to implement following functions:

RadiiTable

libs.RadiiTable uses file assets/Radii.csv, comes from https://pubs.rsc.org/en/content/articlelanding/2008/DT/b801115j

Workflows

Write workflows in a YAML file.

Conecpts

Sections

version

# Syntax version, default is "1" and now must be "1"
version: "1"

rootDirectory

# Root directory for loading files
# relative (from working directory, where command execute) or absolute
# all relative paths below will be walk from this directory
# default: "./"
rootDirectory: "./"

substitutes

# list of directories stores substitutes files
# Directory in the left have higher priority
# Will implicitly use subsitutes in built-in library
# default: "[]"
substitutes: []

template

Load a mol2 file as template

# Specify the template mol2 file. Must provide.
template: "./template.mol2"

jobs

jobs is a sequence of workings to do based on the template.

Each item in jobs should contains follow properties:

name(optional): The name of the item, must be unique.
use: The name of the job
with: Configuration for the job

runners

jobs in workflow are processed with runners. Runners will process each item passthrough the workflow.

A runner function will be built when a job is start with given parameters. The first parameter is options comes from with section in a job, second parameter is meta which contains version, rootDirectory and substitutes specified in global of the yaml.

The runner shuold receive a StaticLayer and a tags(generated by jobs before) as parameters, and returns a processed StaticLayer and a string as tag. Return a list of (StaticLayer, str) is also accepted, while the runner should registered with need_flat as True.

To create a custom runner, you could use Python class like this:

class AddAHydrogen:
  def __init__(self, options, metas):
    self.options = options
    self.metas = metas
  
  def __call__(self, target, names):
    editable = target.to_editable_layer()
    editable.add_atoms([Atom("H", self.options["position"], "new_hydrogen")])
    return editable.to_static_layer(), "add_hydrogen_at_origin"

When use workflow, register your custom runner like this:

from LME.workflow import Workflow, default_runners
import yaml

runners = default_runners | {
  "add_a_hydrogen": (AddAHydrogen, False)
}

data = yaml.load("some_file", yaml.Loader)
Workflow(data, runners)

In the yaml file, call it with:

jobs:
# some steps
- name: add_hydrogen_to_0_0_0
  use: add_a_hydrogen
  with:
    position: [0., 0., 0.]
# some steps

qihexiang/LME