PREDICTOR
@
3D-QSAR.com
by

Designᴵᵀ-TO-LEAD

2nd Computational Medicinal Chemistry Workshop

It is our pleasure to advise Designᴵᵀ-TO-LEAD: 2nd Computational Medicinal Chemistry Workshop.

The workshop will take place on June 13th – 18th, 2022.

This edition will focus on the Structure-Based applications implemented in the www.3d-qsar.com portal.

Know More Remind Later

Predict the Potency of Your Molecule

Welcome to the 3-D QSAR Database. Here you can find a list of target by ChEMBL IDs. For each target at least one or more molecular interactions fields (MIFs) based 3-D QSAR model(s) have been built by the following procedure:
(1) A dataset was created from SMILES structures and associated to their published biological potencies.
(2) The dataset was splitted into training and test set with a ratio of 80:20.
(3) For each dataset a series of conformational analyses were done to obtain 30 conformation for each method. In the actual version th eprogram Balloon (https://doi.org/10.1021/ci6005646) and python based scripts with RDKit (https://www.rdkit.org/) and openbabel (https://doi.org/10.1186/1758-2946-3-33) were used to make 16 different conformational analyses to a maximum of 480 conformations.
(4) The conformational analyses were automatically aligned by means of 6 different methods using the Shaep program () or python based RDKit scripts. For the alignment a preselection of a 17 different template conformations led to thousands of aligned new datasets.
(5) Each alignement was the used to build CoMFA-like models with 4 different force fields for calculating the molecular interactions fields (MIFs), namely TRIPOS (https://doi.org/10.1002/jcc.540100804), GROMOS96 (https://doi.org/10.1016/S0010-4655(99)00540-8), MMFF94 (https://doi.org/10.1002/(SICI)1096-987X(199604)17:5/6<490::AID-JCC1>3.0.CO;2-P) and Autodock 4.0 (https://doi.org/10.1002/jcc.20634). The CoMFA-like model were build by means of a full reimplemented python script (https://doi.org/10.1007/s10822-019-00231-x and https://doi.org/10.1021/acs.jchemed.0c00117)
(6) All the models were built with the same tools available in https://www.3d-qsar.com.

Select a target in the following list and predict your molecule against it.