What is computer-aided drug design?
Computer-aided drug design uses computational tools and algorithms to study compounds virtually before they are tested in the lab. This can involve modeling interactions between drugs and their biological targets, predicting drug properties (e.g., binding affinity), designing structural modifications, and much more. Computer-aided, or in silico, drug design can enable and expedite hit identification, hit-to-lead, and optimization of ADME and toxicology profiles, as well as anticipate safety issues. Our CADD capabilities are available as part of integrated hit identification, hit expansion, hit-to-lead, and lead optimization programs. Computer-aided drug design services are also available as a stand-alone offering.
Computer-Aided Drug Design in Medicinal Chemistry
By helping to guide and accelerate key medicinal chemistry tasks, CADD expedites the design-make-test-analyze (DMTA) cycle, and thereby the delivery of novel therapeutics to the clinic. Medicinal chemistry tasks optimized through the application of CADD include:
- The rationalization of SAR (structure-activity relationships)
- The prioritization of novel compounds proposed by medicinal chemists
- The proposal of novel compound designs to solve project-level issues
Applications of Computer-Aided Drug Discovery
CADD techniques can be applied across the drug discovery pipeline to expedite hit discovery, lead selection, and compound property investigations.
Structure-based drug design
Traditional structure-based drug design uses known structural biology information to identify novel molecules that have the desired shape and charge to bind to the ligand-binding pocket of a known protein structure.
In addition to docking simulations, our panel of advanced techniques includes molecular dynamics (MD), free energy perturbation, quantum mechanical calculations, and water network analysis. With this array of tools, the ligand-target complex can be studied at different levels and, potentially, beyond the static description of x-ray structures, which typically provide a single “snapshot” of a highly mobile and complex system. Our CADD team exploits all available structural information of a target protein to provide you with new hits for your protein target.
Ligand-based drug design
When a target protein’s structure isn’t known, ligand-based drug design uses chemical structure information from known ligands to develop new chemical entities. Techniques such as pharmacophore modeling, 2D and 3D similarity searches, scaffold-hopping, library enumeration, and AI-based generation of new compounds are enabled by in-house tools, allowing for the efficient exploration of a wider chemical space starting from a selection of active hits using industrial and proprietary tools.
Virtual screening
In silico screening involves the virtual docking of potentially millions of chemical structures into a protein structure. Our virtual screening campaigns frequently lead to the rapid and cost-effective identification of excellent starting points for hit-to-lead chemistry. Given one or more compounds with proven activity and/or an X-ray structure of the target, computer-aided drug design and medicinal chemistry scientists apply the appropriate computational techniques in conjunction with in-house and/or external compound collections to identify a set of molecules for biochemical screening.
Protein structure modeling
Our resources allow for the study of challenging targets with limited or incomplete structural data. Whether the modeling of flexible loops or the generation of full-length structures is required, our approaches ensure the generation of reliable models, validated with state-of-the-art computational methods, including molecular dynamics.
Machine learning and predictive modeling
It is critical to exploit all types of experimental information generated on molecules during the discovery process. The computer-aided drug design team has access to software and tools that exploit up-to-date experimental knowledge to generate mathematical models mapping chemical structures to certain properties such as biological activity. Those quantitative structure-activity relation (QSAR) models rely on different approaches, included artificial intelligence, and can be used predictively on new compounds before committing to expensive experimental tests.
High-throughput screening and data triage
Our computational chemistry group works closely with our high-throughput screening (HTS) team to provide computer-aided medicinal chemistry input to HTS campaigns. This includes the selection of compound subsets for screening and can be carried out at either a compound or plate level using various computational methods in combination with our extensive in-house compound collections.
Additionally, after the screen is complete, the computer-aided drug design team assists with data triage using a range of cheminformatics and visualization tools. We can also apply our proprietary hit expansion toolkit to identify additional compounds for screening that can enhance SAR and lead to the discovery of new scaffolds.
Benefits of Computer-Aided Drug Design
Quick Turnaround
Computer-aided drug design technologies expedite the generation of hits and the selection of leads.
Reduced Costs
Computer-aided drug design technologies reduce costs in the early stages of drug discovery.
Connecting with CADD: X-Ray Crystallography
Read about the alignment of computational chemistry and structural biology techniques in this Q&A with our experts.
Get the Insights
Frequently Asked Questions (FAQs) About Computer-Aided Drug Design and Computational Chemistry
-
How do you optimize chemical hits and leads through computational chemistry techniques?
Our computational chemistry team works in partnership with medicinal chemists to drive the optimization of hit and lead series. Our experts will assess each project individually and apply the best computational chemistry methodologies to advance the chemical series towards its target product profile. Structure-based, ligand-based, and predictive modeling will be considered and the most appropriate methodologies selected to maximize the impact on the project.
-
Does Charles River offer QSAR and QSPR modeling services?
Yes, our CADD in medicinal chemistry scientists have the tools and the expertise to advance the various types of experimental information that are generated on molecules during the discovery process. These data are typically classified as "activity" such as IC50 or Ki values from a biochemical assay, or "property," such as physicochemical or in vitro ADMET measurements, giving rise to quantitative structure or property relationship modeling, respectively. The aim in both cases of computer-aided drug discovery is to use the available data to generate mathematical models relating the chemical structure of a molecule to the biological activity or molecular property of interest. The resulting computer-aided drug design models can then be used predictively to help prioritize compounds for synthesis and testing, saving time and resources.
-
Does Charles River leverage artificial intelligence-based methodologies?
Yes, we employ computer-aided drug design software that leverages artificial intelligence (AI) to drive decisions from hit identification to lead optimization. When pursuing AI-enabled drug discovery with us, clients can expect shorter timelines and the highest quality drug candidate molecule.
-
Does your computer-aided drug design team assist with compound library design?
Yes, our computer-aided drug design team has a long history of contributing to the design of compound libraries targeting gene families. More recently, the computer-aided drug design group has assisted in the selection of various sets of fragments that we offer to clients for fragment-based screening campaigns.
We can help clients build or increase their compound collections by selecting compounds from the ever-increasing number of commercially available samples subject to various constraints relating to molecular diversity and/or predicted molecular properties. Compound selections can be additionally focused using protein-ligand docking, molecular similarity ranking, or Bayesian activity models.
-
How does Charles River perform virtual screening?
Virtual screening is a computational method that filters large compound libraries to identify a smaller set of compounds for experimental testing. Virtual screening is an effective computer-aided drug discovery approach. Virtual screening can include ligand-based 2D/3D, identifying likely binders based on structural similarity to known ligands, and structure-based, using protein structure to assess likelihood of binding.
-
Which software is used in computer-aided drug design?
We employ many different brands and types of computer-aided drug design software, depending on client needs and objectives. Charles River is committed to researching and investing in high-quality, best-in-industry computer-aided drug design software for computer-aided drug discovery projects.
-
What is QSAR in drug design?
Quantitative Structure Activity Relationship (QSAR) is a branch of drug design that aims to find correlations between the molecular structure and the observed biological effect that the molecule induces. It is not something specific of computer-aided drug design, but it is possible to use computational resources, including AI for instance, to find patterns within large databases.