Description | Requirements | Installation | Tutorials | Citation | License | Information | Disclaimer
A consensus pharmacophore is a set of properties shared by several active molecules that bind to the same target. It is composed of geometric elements such as points, spheres, vectors, or planes that represent different types of features such as hydrophobic regions, hydrogen bond donors or acceptors, aromatic rings, or positive or negative charges. It can be used to represent the fundamental properties of a molecular interaction and to direct the development of new compounds with comparable or improved activity.
The consensus pharmacophore as Ligand-based method for examining the chemical structures of known active compounds in order to determine the common features that account for their activity. Using the three-dimensional structure of the target protein and its bound ligands, it can provide information about the interaction site and extract the key features required for binding.
A consensus pharmacophore can also be used to identify new potential ligands that match the features of the target and are likely to bind to it. This technique is known as pharmacophore matching, and it is useful for identifying drug targets and performing virtual screening.
This library was developed to generate concensus pharmacophores from large datasets of ligands and ligand-protein complexes.
Question about usage or troubleshooting? Please leave a comment in the discussion section of this repo
ConPhar is reliant on a variety of academic software:
-
pandas
-
pymol
-
plotly
-
seaborn
-
scikit-learn
pip install conphar
If you use this software or its results in your research, publication, or project, please cite it as follows:
If you use pharmer for your work you must cite:
Koes, D.R., & Camacho, C.J. (2011). Pharmer: Efficient and Exact Pharmacophore Search. Journal of chemical information and modeling, 51 6, 1307-14 .
If you use Pharmit in your work you must cite:
Sunseri, J., & Koes, D.R. (2016). Pharmit: interactive exploration of chemical space. Nucleic Acids Research, 44, W442 - W448.
- Fixes
Handle case where a descriptor group has only 1 point in compute_concensus_pharmacophore.__compute_cluster, by setting cluster to 1. Previously, an error would also occur with 2 points, but this is fixed by using the pairwise distance directly (see Changes).
Don't divide by 2 in the radius calculation in compute_concensus_pharmacophore.__compute_center_of_mass_and_radius, as the distance from the furthest point to the center of mass already is the radius.
Only keep alternate conformation A in fetch_structure. Previously, the alternate conformation target_chain was kept, but these are unrelated.
- Changes
Save fetched receptors to a separate folder pdb in fetch_structure
Keep solvent and inorganic within the binding site in fetch_structure
Add indenting to saved pharmacophores in save_pharmacophore_to_json
Use pairwise distance instead of distance of distances in cluster calculation in compute_concensus_pharmacophore.__compute_cluster
Change the threshold on clustering from h_dist * dm.max() to just h_dist (default value adjusted from 0.17 to 1.5)
Remove weighting in center of mass calculation in compute_concensus_pharmacophore.__compute_center_of_mass_and_radius
The radius calculation in compute_concensus_pharmacophore.__compute_center_of_mass_and_radius now adds the radius of individual points, so the full spheres are included in the consensus radius.
Replace balance calculation by variance of the points within consensus clusters
- Additions
Added optional argument target_file_type to fetch_structure to allow fetching mmcif files instead of pdb.
Added optional argument remove to fetch_structure to supply additional residue names to be removed in the preprocessing
Added optional argument feature_size to compute_concensus_pharmacophore to control the default radius per descriptor type
Added optional argument method to compute_concensus_pharmacophore to also allow for different scipy clustering methods
This tool is under MIT license, see the LICENSE file for details.
- This library works with pharmacophores generated with pharmer and/or pharmit. An executable version of pharmit is included in this library but works only for linux.
Users can generate their pharmacophores and use this library for analysis. Check tutorials for more information.
This software is still under development and may contain bugs or errors. The developers do not guarantee the accuracy, completeness, or reliability of the software or its results. Use it at your own risk and discretion. The software is provided "as is" without any warranty of any kind, either express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose, and non-infringement. The developers are not liable for any damages, losses, or costs arising from the use of the software or its results.