Dr. Leandro Radusky and his team have come up with a new online web service, LigQ, that allows fast and efficient identification of potential binders to a desired target before starting a virtual screening procedure. Here, they describe the functionality of the LigQ tool and discuss how you can use it to select and prepare your proteins for virtual screening.
The need for compound selection before virtual screening
A major aspect of drug discovery involves the identification of new compounds that are able to bind a protein and control its activity. In silico virtual screening is one of the most powerful and widely used techniques to search for lead compounds that bind to a protein of interest with moderate to high affinity.
Since the main cost of a virtual screening project is directly related to the number of compounds to be tested experimentally, and given that typically a relatively low number of compounds is selected, it is crucial that this set contains a maxium number of true binders.
The LigQ workflow and pipeline
LigQ is organized into four independent modules that can be used sequentially to perform all virtual screening preparation steps:
1. Pocket Detection Module
This module allows users to find the optimal ligand binding pocket for a given protein target.
2. Ligand Detection Module
With this module users can search a database to find of group of potential binders to the desired protein based on similarity to known binders. Potential ligands are then retrieved and shown on the website as figures or in 3D with JSMol visualizer.
3. Extend Ligand Set Module
From this module, the user can extend the list of compounds found by the previous steps by searching the LigQ database and comparing chemical similarity based on Tanimoto Index.
4. Ligand Structure Generation Module
This module generates enantiomer and tautomer 3D structures for the desired ligands to use them in molecular docking experiments.
An effective and time-saving tool
The pipelined execution of LigQ allows users to start from only a UniProt protein accession to obtain both the docking grid of the most probable binding site of the target protein as well as the candidate compounds in a three-dimensional format ready to execute in silico virtual screening computations.
Since each module can be executed separately, it allows the user to find the most druggable pockets of a protein, a list of compounds with known binding affinity of a protein, an extended set of candidates based on similarity, and the most favorable geometries from a list of compounds.
The LigQ pipeline was also demonstrated to be very effective in retrieving a list of compounds enriched in true binders over commonly used benchmarking sets of proteins.
(Radusky et al., 2017) LigQ: A Webserver to Select and Prepare Ligands for Virtual Screening. Journal of Chemical Information and Modeling.