Computer-aided drug design
Igniting drug discovery
Whether you need to crunch big data, formulate and test hypotheses or simply decide which compound to make for your next iteration, computational sciences are invaluable. As computing power and the quality of algorithms continues to increase, so does the impact of computer-aided drug design (CADD) across all aspects of pharmaceutical development.
Our computational drug designers apply their expertise in informatics and in-silico modeling to fast track your drug discovery journey to the clinic. We combine cutting-edge commercial and open source software with data mining and informatics, and work together with our laboratory scientists to enable:
- Target assessment and modeling
- Virtual HTS
- Structure-based drug design
- Ligand-based drug design
- Lead optimisation
Our CADD experts work with you, as an extension of your team, to advise on the best approach to overcome the scientific hurdles your project is facing. We will discuss your project data and carry out targeted literature searches to help formulate scientific hypotheses and select the appropriate computational tool to test and answer that hypothesis. You can request this service to answer a specific question (as a Fee for Service) or fully integrate our CADD expertise into your drug discovery program.
We provide full reports, models, and data in a range of electronic formats, according to your capabilities and access to software.
For more information about our in-silico drug discovery services, or to discuss how we can accelerate your programs, please contact us.
Target assessment and modeling
Target identification and validation are arguably some of the most critical stages in drug discovery. Our scientists combine in-silico methodologies with in-vitro biology experimentation to evaluate the druggability of your biological targets. Our in-silico target assessment and modeling services include:
- Homology modeling: We build homology models when no suitable protein structure is available. Our models are rigorously refined, trained and validated by our experts. These can be used for hit identification activities (e.g. virtual HTS) and to enable small molecule modulator design in hit to lead and lead optimization
- Binding pocket assessment: We harness proven algorithms to assess the ligandability of your biological target by predicting and ranking binding sites (cryptic, transient, allosteric, etc.) likely to be involved in eliciting the desired biological response
- Selectivity analysis: We combine protein sequence analysis, structural modeling and cheminformatics to enhance target selectivity
Virtual high-throughput screening
Virtual high-throughput screening (Virtual HTS) is a cost-efficient hit identification strategy involving the screening of virtual compound collections against ligand-based or structure-based in-silico models. Our computational scientists will design and execute a bespoke virtual HTS workflow that best suits your needs, and typically includes:
- Virtual library selection: We have a curated virtual collection of nearly 10 million lead-like and drug-like commercial compounds. Our computational scientists can also enumerate virtual compound collections tailored to your needs
- Establish virtual models: We utilize ligand-based or structure-based models. All our models are refined, trained and validated by our modeling experts to ensure fitness for purpose
- Virtual HTS: Virtual libraries are screened in one or more of the bespoke models within a matter of days
- Virtual hit prioritization: A thorough cheminformatic filtration tailored to the biological target class, therapeutic area and specific project needs is combined with clustering to highlight the highest quality virtual hits
Our biology and ADMET screening groups have decades of experience in designing bespoke and efficient screening cascades to confirm and prioritize hits in-vitro, sending your project on the fast-track to success.
In addition to virtual HTS, Concept Life Sciences medicinal and computational chemists have a proven track record in successfully implementing knowledge-based and fragment-based technologies for hit identification.
Structure-based drug design
Knowing the three-dimensional structures of target proteins can have a tremendous impact on your drug discovery process. When suitable structures are available, our computational scientists can use a structure-based drug-design approach, including:
- Binding site analysis
- Molecular docking (reversible, induced-fit, covalent, macrocyclic, etc.)
- Fragment elaboration from fragment screening hits
- PROTAC design for protein degradation
When a receptor structure is unavailable, we work with our X-ray crystallography partners to generate one. We also have extensive experience in building homology models from related proteins structures and applying ligand-based drug design strategies that maximize SAR data.
Ligand-based drug design
No target structure available? Not a problem! Our computational and medicinal chemistry experts will select and apply ligand-based strategies to help you identify high-quality scaffolds for your projects.
- Scaffold hopping and bioisostere scouting: Whether you need to gain intellectual freedom to operate or eliminate liabilities associated with your chemical series, our scientists can help. Our computational scientists can enumerate > 10,000 scaffold hop opportunities, evaluate and prioritise each scaffold hop in-silico individually
- QSAR: We employ field-based QSAR modeling to define rules predicting the activity of novel compounds
- Pharmacophore modeling: We exploit ligand SAR to building lean models defining the spatial arrangement of chemical features that drive receptor recognition. These models not only enable virtual HTS in the absence of receptor structures, but also help identify novel biologically active chemotypes
Our medicinal chemistry team has a proven track record of designing and implementing synthetic routes to the most challenging compounds, thus unlocking novel IP spaces.
Successful lead optimization is a resource-intensive process that aims to identify potent, selective, efficacious and safe compounds suitable for preclinical evaluation.
We strive to run efficient lead optimization campaigns by taking the utmost care in prioritizing the best target ideas. At this stage, each new target we design and synthesize aims to test a specific hypothesis or address drug discovery liabilities. This is often enabled by advanced ab-initio quantum mechanical calculations, including:
- QM pKa determination
- Conformational/rotational energetics
- Tautomeric equilibria
- H-bond strength
- ADME/Tox profiling
- Free energy perturbation
By integrating advances computational modeling, our drug discovery experts will make giant scientific leaps for your projects, saving you months of unnecessary research work.
Our computational scientists employ a range of cheminformatics techniques to support all aspects of drug discovery.
New ideas are generated in-silico (e.g. library enumeration, hit expansion design, 2D- or 3D-similarity searches) and refined by computational and medicinal chemists.
Virtual profiling of these ideas is routinely carried out via the computation of CNS-MPO scores (for CNS indications), pKa and ADMET predictors, and many more.
High volumes of data can be exploited by our experts through molecular matched pair analysis (MMP analysis) and clustering.
We also take pride in constantly aiming for the highest level of data integrity. All our experiments are recorded using our internal electronic laboratory notebook (ELN) systems. We can even work on your ELN platform for a full integration within your systems.