Skip to main content

SPCI: structural and physicochemical interpretation of QSAR models

Project description

SPCI

Automatic tool for mining structure-property relationships from chemical data sets

Description

Retrieves structure-property relationship from data sets in a chemically meaningful way.
Returns estimated contributions of fragments to the investigated property of compounds from a data set and can estimate contribution of different physicochemical factors as well.

Installation

pip install spci

Features

  1. Easy to use straightforward workflow with GUI.
  2. Automatic model building and cross-validation.
  3. Build models for imbalanced data set using the multiple oversampling approach.
  4. Prediction with built models.
  5. Several fragmentation schemes to compute fragment contributions of:
  • common functional groups and rings;
  • Murcko scaffolds;
  • user-defined fragments;
  • automatically generated fragments (based on SMARTS pattern matching broken bonds);
  • per atom fragmentation.

Visualization and analysis of results

  1. Built-in visualization.
  2. rspci - R package for custom visualization and analysis (https://github.com/DrrDom/rspci)
  3. Online tool for visualization, plot customization and figure downloading (http://158.194.101.252:3838/spci-vis/). Demo version is here (http://158.194.101.252:3838/spci-vis-demo/)
  4. Per atom contributions can be visualized with RDKit similarity maps.

Manual

The short manual is included.

Citation

  1. Polishchuk, P. G.; Kuz'min, V. E.; Artemenko, A. G.; Muratov, E. N., Universal Approach for Structural Interpretation of Qsar/Qspr Models. Mol. Inf. 2013, 32, 843-853 - http://dx.doi.org/10.1002/minf.201300029 - structural interpretation.
  2. Polishchuk, P.; Tinkov, O.; Khristova, T.; Ognichenko, L.; Kosinskaya, A.; Varnek, A.; Kuz’min, V., Structural and Physico-Chemical Interpretation (SPCI) of QSAR Models and Its Comparison with Matched Molecular Pair Analysis. J. Chem. Inf. Model. 2016, 56, 1455-1469 - http://dx.doi.org/10.1021/acs.jcim.6b00371 - integrated structural and physicochemical interpretation.

Home page

http://qsar4u.com/pages/sirms_qsar.php

License

LGPLv3

What's new

1.0.0 (03.07.2018)

  • RDKit is used as a backend instead of Indigo
  • multiple undersampling was implemented
  • changed default descriptors, that make this version incompatible with previous models and vice versa.
  • updated sirms descriptors
  • many small fixes and improvements

1.1.0 (07.02.2021)

  • added support of RDKit descriptors
  • added per atom fragmentation
  • reorganized as a Python package
  • console scripts have prefix spci_*

1.1.1 (23.03.2021)

  • changed license to LGPLv3
  • fixed arguments in scpi_descriptors

1.1.2 (28.06.2023)

  • add max_size argument to find_frag_auto_rdkit.py to limit maximum size of output fragments
  • skip fragments with H as a context from output of find_frag_auto_rdkit.py
  • update README and installation notes

1.1.4 (04.11.2024)

  • fix calculation of fragment contributions if initial molecules do not contain hydrogens

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

spci-1.1.4.tar.gz (47.4 kB view details)

Uploaded Source

Built Distribution

spci-1.1.4-py3-none-any.whl (55.9 kB view details)

Uploaded Python 3

File details

Details for the file spci-1.1.4.tar.gz.

File metadata

  • Download URL: spci-1.1.4.tar.gz
  • Upload date:
  • Size: 47.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.0.0 CPython/3.10.8

File hashes

Hashes for spci-1.1.4.tar.gz
Algorithm Hash digest
SHA256 2325a506b812991244a170dcaede4b6dcb50a5935f1f1e181fe6e407e7aa5e28
MD5 5495c07ab3aebf377e4c8d76941e48c7
BLAKE2b-256 ce15885ebd7b398b7d491ddec994087220fb4e3c529b9f2b3f8412a284e3375a

See more details on using hashes here.

File details

Details for the file spci-1.1.4-py3-none-any.whl.

File metadata

  • Download URL: spci-1.1.4-py3-none-any.whl
  • Upload date:
  • Size: 55.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.0.0 CPython/3.10.8

File hashes

Hashes for spci-1.1.4-py3-none-any.whl
Algorithm Hash digest
SHA256 2cba16fa9585cc355db7d614db5dd096d1d188fb2825b15c411ea7afa6725517
MD5 28938d87f1e91c9f7af9a19687efd7e3
BLAKE2b-256 1facab5307425262cd0d66faf403300b8bff2025e3147b06d1e83f9370d23b44

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page