lserest 1.1.2

Provides access to Leadscope Web Services REST API via Python.

Leadscope Python Interface

Provides access to Leadscope Web Services REST API via Python. This is an interface client which requires access to a Leadscope Enterprise server, v2023.0.1 or later.

Getting started

Install the package:

pip install lserest

Server connection details should be set in config.toml (see below for format) and placed in working directory, or specified: LSEClient(config_file=<path>). Alternatively, some or all of the parameters root_url, username, password and company may be passed directly to the constructor.

Config file

The config.toml file (if used) should contain the following section:

[server]
root_url = "http://server_host_name:8080/rs/resources"
username = "<user>"
password = "<password>"
company  = "<company>"

Server sessions

Most server requests require an active session to be created via LSEClient.start_session. Such sessions must be closed after use using LSEClient.end_session. A preferable option is to use LSEClient.with_session, which creates a session and then closes it at the end, handling any exceptions which may be thrown, all examples will use this method. See the documentation for the classes LSEClient, StructureSet, StructureSearch, ModelApplication and NitrosamineAssessment for a full list of which methods require a session to be open.

Examples

Requesting a list of available statistical models from the Leadscope server:

from lserest import LSEClient, StructureSet, ModelApplication
client = LSEClient()

def example1(c):
    c.list_models()

client.with_session(example1)

Running a model application, specifically the ICH M7 Consensus, on two structures:

from lserest import LSEClient, StructureSet, ModelApplication
client = LSEClient()

def example2(c, molfiles):
    structures = StructureSet(c, molfiles)
    model = ModelApplication(c, structures, consensus_names=['Genetox'])
    model.apply()
    return model

model = client.with_session(example2, ['structure1.mol', 'structure2.mol'])

print(model.consensus_results())

Searching the database for similar molecules to a given structure, downloading structural images of each search result and saving them to a zip archive:

from lserest import LSEClient, StructureSet, StructureSearch
import numpy as np
client = LSEClient()

def example3(c):
    structures = StructureSet(c, molfile)
    search = StructureSearch(c, structures)
    search.similar(match_pct=70)
    c.save_structure_images(np.unique(search.results()['Structure ID']), 'search_results.zip')

client.with_session(example3, 'structure.mol')

Running an N-nitrosamine potency assessment, on two structures:

from lserest import LSEClient, StructureSet, NitrosamineAssessment
client = LSEClient()

def example4(c):
    structures = StructureSet(c, 'structure1.mol', 'structure2.mol')
    nitro = NitrosamineAssessment(c, structures)
    nitro.apply()
    return nitro

nitro = client.with_session(example4)
print(nitro.results())

Creating a StructureSet from mol strings:

mol_string = '<mol string>'
from lserest import LSEClient, StructureSet, NitrosamineAssessment
client = LSEClient()
structures = StructureSet(client)
structures.mol_strings.append(mol_string)