University of Warwick and Cresset Drug Design Roadshow

We are excited to announce that we will be hosting the inaugural University of Warwick and Cresset Drug Design Roadshow on Wednesday 6th and Thursday 7th September 2023.

This two-day interactive event will emcompass several areas within the field of computational chemistry, and will provide a flexible learning platform for attendees to experience Cresset software.

Wednesday 6th September will host a series of lecture style sessions, followed by hands-on software workshops on Thursday 7th September. Attendees are encouraged to attend both days of the event to experience the full benefit of the event.

Please find registration at the bottom of the page.

Who should attend?

Registrations are welcome from all those working in life sciences research, particularly encouraged from University of Warwick and surrounding local Universities, including industry and academic spin-outs. Academics, Postgraduates students, Undergraduate students, research scientists and technical staff are all encouraged to join to enhance their knowledge, or begin their educational journey within computational chemistry.

Agenda

Wednesday 6th September

Abstracts

An introduction to computational chemistry and its application in life science research
Dr. Rob Scoffin, Chief Executive Officer, Cresset 

Computational methods are a powerful tool in life science research, enabling teams to increase research efficiency and remove unnecessary wet lab work. Starting at the most basic level, we will introduce computational chemistry methods and how they are applied in life science research.

Identifying novel inhibitors against tuberculosis
Dr. Nathan Kidley, Principal Application Scientist, Cresset 

Fragment-based drug discovery is an evolving area of research with unique challenges in developing the hits into lead molecules. In this case study we use the bioisostere solution, Spark™, to demonstrate how structure-based virtual screening effectively identifies novel InhA reductase inhibitors for tuberculosis (TB) therapies. We demonstrate how the docking score capabilities in Spark are used to grow a fragment into a new region of the active site to identify new areas of chemistry.

Penicillin binding protein inhibitors of EcPBP3 discovered using a modified phage display platform (Bicycles®)
Dr. Cathy Rowland, Principle Scientist, Bicycle Therapeutics 

Bicycle Therapeutics is developing a unique class of chemically synthesised medicines based on its proprietary bicyclic peptide (Bicycle®) phage display platform to address therapeutic needs unreachable with existing treatment modalities. In addition to its internal focus in oncology, Bicycle has secured external funding to investigate the application of the novel platform to the field of antibacterials, an area of increasingly high unmet medical need. Cyclic peptide moieties are wellrepresented amongst successful antibiotics but have mostly been discovered through serendipity from natural sources by laborious screening processes. We present a proof-of-concept in application of the Bicycle platform to antibacterial targets, specifically the discovery of novel inhibitors of penicillin binding proteins (PBPs). Peptides of interest from selections against Escherichia coli PBP3 were chemically synthesised and characterised using fluorescence polarisation, MIC and cytotoxicity assays. We describe a Bicycle peptide with highaffinity binding of E. coli PBP3 and a viable spectrum of killing activity against clinically relevant Enterobacteriales species. Our Bicycle peptide represents a novel non-covalent inhibitor of E. coli PBP3.

Development of non-opioid analgesics via cannabinoid CB2 modulation
Dr. Scott Midgley, Application Scientist, Cresset

Development of non-opioid analgesics is a key interest in modern pharmaceutical sciences due to the increasing frequency of opioid prescription and attenuated addiction. Targeted Cannabinoid receptor modulation is a promising alternative to opioid-based analgesia; however there are significant technical challenges in developing selective CB2 agonists (analgesics) while avoiding the psychoactive effects that result from ligand-binding to CB1.

In this talk, we will present a case study based on virtual screening and triaging of compounds using Cresset’s CADD software environment. We show that a combination of Cresset tools combined with expert medicinal chemistry knowledge from Cresset Discovery can generate novel and promising leads for modern drug discovery workflows.  

External perspective on Artificial Intelligence
Dr. Gregory Ross, Research Leader, Isomorphic Labs

It is well recognized that computational techniques can speed up the identification of hits and accelerate the optimization of such hits to lead series and development candidate molecules. A class of rigorous physics-based methods known as binding free energy (BFE) calculations have emerged as the most consistently accurate relative affinity prediction tool available to support such efforts. Yet, there remains uncertainty about how accurate these techniques are and how accurate they can ever be. This presentation will investigate the prediction accuracy of a leading BFE method on the largest publicly available dataset of proteins and congeneric series of small molecules. The limit of achievable accuracy for any prediction method is interrogated using the results from a survey on the reproducibility of experimental relative affinity measurements. The implications of these results for future BFE methods will be discussed.  

AI-based generation of custom torsion parameters for small molecules
Dr. Venkat Ramaswamy, Senior Computational Chemistry Developer, Cresset

Accurate force field parameters are essential for reliably predicting the thermodynamic properties of small molecules in Molecular Dynamics (MD) and Free Energy Perturbation (FEP) calculations. Torsion parameters, in particular, are crucial as they describe the largest local motions within the molecule, thus impacting the ligand’s conformational distribution in these calculations.

Traditionally, custom torsion potentials are generated from quantum mechanical (QM) rotation scans, which involve significant computational cost depending on the flexibility (number of rotatable bonds) of the molecule and the accuracy of QM description used. In this presentation, I will discuss how our implementation of ANI-2X - a machine learned potential specifically refined to better predict torsion profiles - provides the accuracy of its reference density functional theory (ωB97X/6-31G(d)) at a fraction of the computational cost.

MEMENTO:  Why it is useful to be forgetful in conformational sampling
Prof. Philip Biggin, University of Oxford

As part of their function many membrane proteins, such as transporters, transition between multiple states. Having representative structures of the distinct conformational states and how they transition between them would be extremely useful not just for furthering our understanding of key biochemical processes but also for the rational design of modulatory drugs. However, structures are only determined in free energy minima, one at a time. Furthermore, whilst conformational flexibility can be inferred from static end-state structures, their interconversion mechanisms are often beyond the scope of direct experimentation. On the other hand, molecular dynamics (MD) offers a highly complementary technique that is very appropriate for the study of conformational transitions and indeed there have been many studies reporting such attempts.  However, ensuring convergence and reversibility in such transitions is extremely challenging. A typical approach is to generate a pathway between two end-states via the use of steered MD (SMD) and then employ umbrella sampling to explore the energetic difference between the two states along the pathway. However, this approach can suffer heavily from starting-state dependence (hysteresis) and the resulting energetics can be dramatically incorrect.

In this talk, I will discuss how we recently addressed this problem through a new history-independent approach that we call “MEMENTO” (Morphing End states by Modelling Ensembles with iNdependent TOpologies) to generate paths that alleviate hysteresis in the construction of conformational free energy profiles. I will compare SMD and MEMENTO on well-characterized test cases (the toy peptide deca-alanine and the enzyme adenylate kinase) before discussing its use in more complicated systems (the kinase P38α and the bacterial leucine transporter LeuT). The approach should facilitate the study of conformational changes in traditionally difficult problem proteins and how these transitions are influenced by both orthosteric and allosteric compounds.

Using Flare™ FEP to identify novel inhibitors for CDK9 and SARS-CoV-2 Mpro
Dr. Sofia Bariami, Associate Product Manager, Cresset

In this presentation, we discuss Free Energy Perturbation (FEP) calculations which can deliver reliable and robust predictions of relative binding affinities for a congeneric series of small molecules. This is done by considering the relative differences in chemical structure of the compounds as well as the interactions being made with the target protein. The insights obtained through FEP calculations can help researchers prioritize the most promising compounds worthy of further experimentation in the lab.

Flare FEP combines efficient network generation, adaptive lambda sampling and custom force field parameter tools to facilitate rapid and robust FEP calculations. Through two case studies, we will showcase the practical application of Flare FEP in identifying novel inhibitors.

The first case study considers searching for inhibitors of cyclin-dependent kinase 9 (CDK9), while the second example explores the targeting of SARS-CoV-2 MPro. By employing Cresset's CADD software, we will demonstrate how to navigate the chemical space effectively, ultimately discovering ligands with significantly enhanced binding free energy compared to the initial molecules.

Posters

Poster and networking sessions will be held on Wednesday 6th September to give the opportunity for attendees to showcase their research. Prizes will be given to selected, outstanding posters based on the decision of the reviewing commitee. 

If you will be presenting a poster, please include your poster abstract during registration (maximum 300 words). Alternatively, please email this to katie.nicoll@cresset-group.com by Wednesday 30th August.

Workshops

Thursday 7th September

Hands-on software workshops will run in two parallel sessions. Attendees will be provided with a 90-day free license to Flare™ and Spark™ ahead of the workshops.

Please note that there are limited spaces for these workshops and spaces will be allocated based on computational chemistry experience and on a first-come-first-served basis. 

Please find registration at the bottom of the page.

Location

University of Warwick, Coventry CV4 7AL

Registration

This event has now finished. Please look out for upcoming webinars and events on our website.