This meeting is primarily designed to address the needs of Cresset software and consulting services users, however, non-users are welcome to attend. The content of the meeting will enable you to understand more about how Cresset solutions can increase the effectiveness of your drug discovery programs.
In our industry, where confidentiality is so important, it is a mark of respect for each other that we manage to share the inspiration and ideas behind our common scientific goals in a constructive way. As has been the case at previous Cresset user meetings, we once again look forward to the open sharing of ideas, approaches and results.
|9:00||Registration and networking|
|9:20||Welcome||Rob Scoffin, Cresset|
|9:30||Summarizing activity and selectivity in one picture||Mark Mackey, Cresset|
|10:00||Computational approaches to the new discovery of new CRAC channel antagonists||Dave Madge, Calcico Therapeutics|
|10:30||What’s new and different in Cresset software||Tim Cheeseright & Giovanna Tedesco, Cresset|
|11:00||Break and networking|
|11:30||Field based analysis of kinome ligand space||Marcin Krol, Selvita|
|12:00||Optimization of protein kinase inhibitors – how structural information can help||Daniel Kuhn, Merck Serono|
|12:30||Predicting agonism in GPCRs||Martin Slater, Cresset|
|1:00||Lunch and networking|
|2:10||Should I sow QM in my fields?||Ewa Chudyk, Evotec|
|2:40||Spatial overlap of peptide hotspots and canonical drug pockets in a model enzyme||Walraj Gosal, Isogenica|
|3:00||Fragment-based screening, what can we learn from published hits?||Chris Swain, Cambridge MedChem Consulting|
|3:20||Analyzing protein active sites using the XED force field||Mark Mackey, Cresset|
|3:50||Break and networking|
|4:20||Practical experiences of using the 3D-QSAR tools in Forge||Bohdan Waszkowycz, Cancer Research UK|
|4:50||Identification of efficacious bioisosteres using the field based approach in conjunction with orthogonal methods||Zara Sands, UCB|
|5:20||Large-scale compound clustering in 3D||Paolo Tosco, Cresset|
|5:50||Closing remarks||Rob Scoffin, Cresset|
|6:00||‘Magical’ drinks reception|
|9:00||Registration and networking|
Introduction to Cresset software for medicinal / synthetic chemists
Computational chemistry workshops
Computational chemistry workshops
|9:30||Scaffold hopping for new ideas with Spark||Virtual screening from the desktop with Forge and Blaze||Explore the new features in Forge|
|11:00||Break and networking||Break and networking||Break and networking|
|11:30||Perfecting your new molecule designs with Torch||Using Cresset in workflows – Spark in KNIME||Tips and tricks in Cresset software|
Mark Mackey, Cresset
3D-QSAR based on molecular interaction potentials can provide a wealth of information about the exact molecular characteristics required for activity. However, current techniques have a number of issues such as alignment noise, sampling errors and descriptor choice which can make it difficult to reliably produce effective models. We have presented in the past techniques for solving the sampling problem and shown that using accurate electrostatics combined with simple shape descriptors often gives meaningful models. However, there are still times when it is not possible to obtain a statistically valid linear regression model.
One useful qualitative data analysis method that is being increasingly used is activity cliffs analysis. In this technique, pairs of compounds are located that are similar (in some sense), but have different activities. Traditionally activity cliff analysis has used a 2D definition of similarity, but extension to 3D similarity metrics gives additional information that is very useful to locate the source of and reason for the activity differences.
An extension of 3D activity cliff analysis is to mine the entire data set for corresponding cliffs and use this to build a model for activity. Analysis of the data set to locate activity cliffs locates the pairs of molecules with the highest information content. However, this needs to be tempered with an analysis of how likely it is that the molecules are aligned correctly, as only properly-aligned molecules contain any information. We apply Bayesian corrections to the activity cliff data to obtain a map of the electrostatic and shape characteristics that seem to locally correlate with improving activity. The resulting model is semi-quantitative in that it attempts to describe the entire data set without building a linear regression model. This technique provides a valuable fall back to the computational chemist for information extraction from ligands in 3D.
Dave Madge, Calcico Therapeutics
The calcium release activated calcium (CRAC) channel is an interesting target for a range of autoimmune and inflammatory disorders for which modulation of T-cells, B-cells and mast cells might be an appropriate therapeutic strategy. Historically this ion channel has been associated with ligands of poor selectivity and DMPK properties, in part arising from the difficulty of establishing high-quality primary assays that can be used to drive SAR. At Calcico Therapeutics we have addressed the deficiencies in primary screening and used this, in conjunction with both computationally-driven screening set selection and traditional HTS, to identify novel ligands. Recently available structure information for this ion channel provides further opportunities for rational design based on key interactions of regulatory components.
Tim Cheeseright & Giovanna Tedesco, Cresset
The latest advances in all of the Cresset applications will be presented including live demonstrations and case studies. They will exemplify much of the functionality that was released in 2014 including the tile view, result tags and identification of selectivity cliffs using Forge. The interfaces to the latest scientific advances will also be demonstrated including our new rapid SAR analysis module.
Marcin Krol, Selvita
There is a vast amount of structural and biochemical data related to kinase inhibitors which can be used to leverage our efforts in the areas of inhibitor selectivity, potency and novelty. Here, an alternative protein kinase similarity space is presented based on the field-based similarity analysis of a broad set of selective and promiscuous kinase inhibitors. Specific structural and field aspects of the resulting clusters of protein/ligand pairs are investigated to gain deeper understanding of the features responsible for inhibitor selectivity. Common features characteristic of the non-specific kinase binding and features responsible for selectivity are also discussed. Additionally, the study helps to identify non-obvious clusters of protein kinases similar in the inhibitor space, which can be used in the definition of off-targets for a kinase inhibitor project. Finally, analysis of functional fragments within clusters of selective ligands may facilitate the design of novel, potent and selective kinase inhibitors.
Daniel Kuhn, Merck Serono
In the optimization of kinase inhibitors towards pre-clinical development multiple parameters have to be fine-tuned in parallel. How can structural information help in this process? The optimization of c-Met inhibitors will be presented as use case. The influence of ligand- and structure-based design techniques will be discussed.
Martin Slater, Cresset
The XED force field, developed by Dr Andy Vinter, is the foundation of all Cresset commercial field based software. Though the software is considered by many as ‘best in class’ for a number of ligand centric computational chemistry workflows, including scaffold hopping, virtual screening and bioisosteric fragment replacement, the XED force field is less well known for its wider utility in modeling larger molecular systems such as proteins. This is in spite of the fact that the Cresset consulting services team have continually used and developed this, over a decade, for both ligand and protein modeling work.
In order to highlight the finesse of the multipole force field, we have set ourselves an extreme challenge which, for some, represents the holy grail in GPCR research: the potential of predicting a pharmacological outcome. The latest GPCR structures have provided a unique opportunity to study the complete GPCR system incorporating both transmembrane protein and effector G-proteins. We have begun probing these systems and, with the resolution that the XED force field can provide, are beginning to glimpse the possibility of unravelling the effects of ligands on the energetics of G-protein binding in a quantitative manner.
The results of Dr Vinter’s pioneering GPCR modeling work will be presented.
Ewa Chudyk, Evotec
Accurate prediction of ligand binding interactions is a key element of drug design. Many of the currently used methods rely on force field based calculations, often presenting a compromise between speed and accuracy. However, due to the increase in computer power in recent years, Quantum-Mechanics (QM) based tools are increasingly accessible for estimating binding enthalpy. The ab initio fragment molecular orbital (FMO) method, using linear scaling to perform MP2 (Møller–Plesset perturbation) calculations, incorporates many non-classical intermolecular forces. Pairwise interaction energy analysis enables us to deconvolute the components of molecular interactions into electrostatic, charge transfer, exchange repulsion and dispersion interactions. Including such interactions for studying protein-ligand binding has been shown to improve structure-based drug design.1
Inspired by the recent attempts of protein field sampling2, here we would like to investigate Cresset field technology in terms of protein-ligand interactions in structure-based drug design. Complementarity of protein and ligand fields will be discussed with MP2-based interaction energies for selected examples. Further analysis of those complexes will also indicate whether QM effects are important for describing protein-ligand binding.
(1) Mazanetz, M. P., Ichihara, O., Law, R. J., and Whittaker, M. (2011) Prediction of cyclin-dependent kinase 2 inhibitor potency using the fragment molecular orbital method. J. Cheminformatics 3, 2.
(2) Comparing the electrostatic properties of protein active sites and other Cresset research. Cresset.
Walraj Gosal, Isogenica
The discipline of drug discovery is often polarised between small molecule and biological drugs. Whilst biologics represent the fastest growing class of drugs, small molecules offer advantages through bioavailability and cost. In this talk, we present data to demonstrate how biologics, discovered by using molecular display rather than high throughput screening, can be used to inform small molecule research and therefore bridge the gap between these modalities. In particular, we are investigating whether we can decorate the surface of proteins with peptide probes which act as structural templates for pharmacophore discovery. We discuss recent structural, in silico and biochemical data to support this concept for novel peptides in complex with thrombin in collaboration with Cresset.
Chris Swain, Cambridge MedChem Consulting
Fragment-based screening has now become an essential tool in drug discovery, there are now a number of different screening technologies employed and a wide variety of diverse targets addressed. This presentation will attempt to address the following questions: Can we design better fragment libraries? Does the technology/target impact the kinds of fragment hits identified? Can fragments help us better understand the way that small molecules interact with proteins?
Mark Mackey, Cresset
The Cresset XED force field has shown itself to be excellent at computing the electrostatic properties of small molecules. One of the goals at Cresset has been to apply the unique features of XED to proteins. We present preliminary results from two research projects in this area: depicting the electrostatic environment in a protein active site and using that to assist in ligand design, and using XED combined with 3D RISM theory to get an accurate picture of the energetics of bound waters in a protein active site.
Bohdan Waszkowycz, Cancer Research UK
Generation of 3D-QSAR models can be a useful technique during lead optimisation to support compound design and to develop improved scoring schemes. However, it is important to assemble a training set with appropriate affinity and diversity, and to evaluate the impact of alternative alignment protocols. During the course of a project to identify selective inhibitors of the tyrosine kinase RET, a target of increasing clinical interest in the management of medullary thyroid cancer and non-small cell lung cancer, we have synthesised a large number of analogues around several distinct scaffolds in order to explore SAR for various regions of the ATP binding site. Retrospective analysis of this dataset using the 3D-QSAR tools within Forge suggests some practical guidelines for generating informative and robust models of inhibitor affinity and selectivity.
Zara Sands, UCB
Designing dual activity in a single molecule can be an onerous task as the available consensus chemical space leading to desirable biological, toxicological and DMPK properties can oftentimes be narrow. I will present a lead chemical series developed in a dual target program which possessed a chemical moiety that was key for dual activity but responsible for genotox issues. I will describe a strategy employing LB (MMP & Torch) and SB (Recore) approaches that we used to identify active and genotox free bioisostere replacements.
Paolo Tosco, Cresset
Clustering a collection of n compounds requires the computation of a triangular similarity matrix filled with pairwise similarity values. Since the cost for computing such a matrix is O(n2), fast similarity metrics based on 2D fingerprints are most often used for this purpose. However, a 2D metric has significant inherent limitations in capturing the biological similarity across conformationally flexible molecules. Large variations in functional group decoration may marginally affect 3D steric/electrostatic properties; conversely, moving from an extended to a folded conformation may have a dramatic influence on recognition by a macromolecular target, which is completely ignored by 2D methods.
We have recently been involved in a collaborative project with BioBlocks, namely clustering their Comprehensive Fragment Library (CFL), a non-random selection of about 800K variably decorated heterocyclic core structures generated from first principles. When 2D ECFP4 fingerprint Tanimoto distances were applied to a subset of this collection, results were disappointing: the distribution of 2D similarity values was rather flat across the set, and there was no correlation with 3D similarity metrics. The largely unprecedented chemical nature of structures and the 3D-oriented design of the library called for high quality, three-dimensional conformer generation, molecular alignment and similarity metrics.
Methodological and technical solutions adopted to enable 3D clustering of such a large collection will be presented. The higher quality and informative content of 3D vs 2D clusters will be illustrated through selected examples.
After a DPhil in Chemistry at the University of Oxford, Tim gained experience as both a medicinal chemist and a molecular modeler at Peptide Therapeutics and Medivir. Tim joined Cresset in 2002. As Director of Products he is responsible for delivering easy to use applications that solve key problems in small molecule drug design and discovery.
Ewa Chudyk gained her PhD at the University of Bristol (UK) in 2013 working in Prof. Adrian Mulholland’s research group. Her PhD thesis was on modeling catalytic mechanisms of class A beta-lactamases (responsible for antibiotic resistance) and fatty acid amide hydrolase (antidepressant and sleep-inducing effects). She has gained a broad experience in modeling chemical reactions using quantum mechanics/molecular mechanics hybrid approach, as well as empirical valence bond methods. During her post-doctoral experience at the Technical University of Munich (Germany), Ewa has applied those skills to important drug targets including immunoproteasome, HCV protease and ATP-dependent Clp protease (antibacterial target). She joined Evotec (UK) computational chemistry group in August 2014, where she collaborates with medicinal chemistry teams to support drug discovery projects. Through her career Ewa has worked on various drug discovery projects, including modeling enzyme catalysis and molecular interactions using ab initio calculations. In her future research she would like to concentrate on developing fast and accurate structure-based drug design tools.
Walraj Gosal is a Senior Scientist at Isogenica, a small Biotech company that is using CIS display and Colibra Library technology to engineer protein biologics and antibodies. Prior to joining Isogenica, Walraj was a postdoctoral fellow at the University of Texas, Southwestern Medical Center and at the University of Leeds in the area of protein folding, design and evolution.
Dr Marcin Król works as a Senior Computational Chemist at Selvita SA a Polish biotechnology company. His main research interests are in the field of oncology drug design with a specific focus on non-kinase targets. He is involved in the early stages of drug development including target selection and druggability analysis, virtual hit finding campaigns and focused library design.
Marcin obtained his PhD at the Jagiellonian University in Kraków, where he worked on the computational studies of immunological signal transduction in antibodies upon antigen binding using molecular modeling techniques.
After completing his PhD Marcin joined Dr Paul Bates’ group at Cancer Research UK as a post-doctoral fellow, where he worked on novel fully flexible protein-protein docking algorithms. The algorithms were successfully used in the community-wide prediction of protein-protein interactions CAPRI. While at Cancer Research UK Dr Król participated in the collaborative project on the computational design of point mutations to increase stability, activity and selectivity of L-Asparaginase, an enzyme used in the treatment of Acute Lymphoblastic Leukemia (ALL).
Marcin has coauthored over 30 papers in peer reviewed journals and several patents.
Daniel Kuhn is currently computational chemist and discovery project leader at Merck in Darmstadt, Germany. He obtained his PhD from Philipps-University, Marburg in computational chemistry working with Gerhard Klebe on the classification of protein binding sites. In 2004 he joined Boehringer Ingelheim in Vienna as computational chemist working in oncology research. In 2010 he joined Merck as senior scientist contributing to projects in early and late drug discovery stages. He is also leading a lead optimization project towards preclinical development. His research interests include protein kinase drug discovery, structure-based design and knowledge-driven approaches to hit identification and lead-optimization.
Mark graduated from the University of Tasmania in 1992 with a BSc in Chemistry, Physics and Mathematics. Between 1988 and 1992 Mark was awarded the following accolades: Chemistry Honors Prize at University of Tasmania; Dean’s citation for outstanding undergraduate and Honors achievements; University Medal from University of Tasmania; ICI Australia Ltd Prize for third year chemistry; Masson Memorial Prize for top chemistry graduate in Australia.
In 1993 Mark moved to the University of Cambridge where he was awarded his PhD in 1997. It was during his time at Cambridge that Mark first worked with Dr Andy Vinter, founder of Cresset.
Upon completion of his PhD Mark worked as a Molecular Modeler and Senior Research Scientist for Napp Pharmaceuticals and Merck Sharp & Dohme respectively.
In 2002 Dr Vinter invited Mark to join him as a founder member of Cresset. Mark has designed and added fundamental science to tools that are now accepted and used by many pharmaceutical, agrochemical and academic institutions across the world. In 2010 Mark was appointed as CSO of Cresset.
Dave Madge recently joined Calcico Therapeutics (Oxford, UK) as VP, Chemistry, and also holds a position as Project Director for WuXiAppTec (Shanghai, China). Dave joined Calcico from Xention, an ion channel drug discovery company based in Cambridge, UK. After joining Xention in 2005, as VP, Chemistry, Dave went on to manage discovery projects in the therapeutic areas of pain, cardiology, urology and respiratory disorders, becoming VP, Research in 2009. Prior to Xention, Dave was Chief Operating Officer of a chemistry services company, NCE Discovery, a spin-out from an academic drug discovery group at University College London, which was later merged into an integrated chemistry and biology CRO called Domainex. Dave spent 8 years at UCL running integrated drug discovery projects in an academic environment, and directing medicinal chemistry activities in spin-out companies. Prior to joining UCL Dave was a medicinal chemist at The Wellcome Foundation, working on exploratory drug discovery projects across CNS and cardiovascular therapeutic areas. Dave gained his PhD at Imperial College London, working on new fluorination techniques for positron emission tomography ligands.
After graduating from Manchester University with a 1st class BSc Hons degree in Chemistry Zara Sands embarked upon a PhD in the Cancer Research School, at Nottingham University where she successfully completed her doctoral training in computational medicinal chemistry. In 2003 she took a Welcome Trust Fellowship position at the University of Oxford and under the direction of Prof Mark Sansom developed an expertise in membrane protein structural biology & biophysics. In 2006 Zara joined AstraZeneca where she applied and developed cutting edge in silico technologies for studying challenging CNS targets. She also served as Complex Systems Modelling Advisor to AstraZeneca’s global computational chemistry strategy team. In 2009, she was recruited by UCB BioPharma to support and strengthen their CNS drug discovery pipeline through the continued application and development of in silico technologies towards membrane protein targets. She has been instrumental in developing UCB BioPharma’s GPCR drug discovery platform and through the judicious application of computational approaches has been able to successfully drive UCB’s orthosteric and allosteric GPCR drug discovery projects.
Dr Robert Scoffin is an experienced entrepreneur and senior executive within the life sciences software industry. Following a DPhil gained from Oxford University, he worked for eight years at Oxford Molecular, seeing the company grow from 15 to 250 people and a successful floatation on the London Stock Exchange.
In 1998 Rob joined CambridgeSoft, at the time a well-known creator of desktop chemistry software, with the remit to help grow an Enterprise Informatics business with particular focus on the European market. He achieved great success with CambridgeSoft over a twelve year period, managing European Operations with responsibility for marketing, sales, support and professional services. Rob oversaw the growth of the business to more than $18m in annual turnover, being responsible for a team of more than 30 direct employees and more than 65 contract staff.
Rob joined Cresset as CEO in 2010. During the time he has been leading the team, Cresset has further developed its software and services from a very solid customer base, into a high-growth and profitable business.
In 2014 Rob co-founded Re-Pharm, an early-stage drug discovery and development company based on the principles of compound re-profiling.
Dr Martin Slater studied medicinal chemistry at the Universities of Leeds and Huddersfield.
In 1997 Martin joined the then start-up company BioFocus where, as Senior Research Fellow, he underpinned the SoftFocus library brand with the development of innovative chemogenomic tools and the design of over 40 commercially successful protein targeted libraries. Martin pioneered the use of Cresset field based technologies for targets including GPCRs, Kinases, Ion channels and Proteases for library generation and de-novo ligand design.
Martin joined Cresset in 2011 as Director of Consulting Services, delivering high quality computational chemistry and molecular modlling service to pharma, biotech, agrochem, academia, flavors and fragrances industries amongst others.
Chris founded Cambridge MedChem Consulting in May 2006. Prior to that he spent nearly 20 years at Merck.
Recent activities include, providing medicinal chemistry support for two projects supported by the ‘Seeding Drug Discovery Initiative’ initiated by Wellcome. Licensing oversight and review of possible licensing candidates. Chris is a member of the Cancer Research UK DDAG and recently joined the Scientific Advisory Board of Selcia Ltd. Ongoing collaborations include multiple small, medium and large Pharma companies and academic groups providing lead finding oversight, lead optimisation and computational chemistry input. Chris also provided training/talks for RSC summer school and Pharma companies.
At Merck Chris lead the NK1 Antagonist Project. Eight compounds selected for preclinical development, to date four have entered clinical development and one has been marketed (Emend). Co-author of background documents and presentations. Winner of the RSC 2000 Biological and Medicinal Chemistry Sector Prize, for work on NK1 antagonists. Chris initiated five projects as the Chemistry Director of the Pain Franchise, all projects now have selected compounds for preclinical development. He also initiated the Basic Research Electronic Notebook project. Representative for Terlings Park on the Chemical Technology Working group, responsible for bringing novel technologies into Merck. Terlings Park representative on the committee responsible for enhancing the Merck sample collection, designed and initiated numerous libraries. Chris was responsible for Terlings Park computational chemistry effort. Built first web-based portal for project information, allowing structure and activity based searching and document/report management. He built HTS_browser a tool for analysis of HTS data, responsible for analysis and initial follow-up of all screens initiated by Terlings Park. Responsible for analysis of the drug discovery process and an analysis of the progression of compounds through safety assessment.
Dr Giovanna Tedesco joined Glaxo in 1990. As senior computational chemist she supported a variety of drug discovery programmes in the antibacterials and CNS areas, and led target-to-lead CNS programmes. When Aptuit took over the GSK site in Verona, Giovanna moved to client services where she worked as a senior proposal driver.
Giovanna joined Cresset in December 2014. As product manager she has responsibility for delivering software for computational chemists.
Paolo completed his PhD in Drug Science in 2002 at the University of Turin (Italy), and in 2004 he was appointed Assistant Professor in Medicinal Chemistry. Starting from 2005 he gradually moved from synthetic chemistry to computational chemistry, which is currently the focus of his research interests. From 2008 to 2013 he released a number of open-source packages dedicated to molecular alignment and 3D-QSAR, and established successful collaborations with both academic (University of Geneva, University of Copenhagen) and industrial partners (COSMOlogic, Novartis Institutes for Biomedical Research). In 2012 he was the winner of the Teach-Discover-Treat challenge, an initiative promoted by the COMP division of ACS to foster the development of drug discovery computational workflows for neglected diseases. In 2014 he joined Cresset as Computational Chemistry Developer.
I have been based in the Drug Discovery Unit at the CRUK Manchester Institute since 2011, providing computational chemistry support on a wide range of oncology targets in areas such as DNA repair, epigenetics and tumour cell metabolism, with a particular interest in the application of structure-based design and virtual screening. Following a degree in Pharmacy and a PhD in QM/MM methods at the University of Manchester, I joined Proteus Molecular Design (later Protherics) in 1990, heading up the computational chemistry group after acquisition by Tularik in 2001, and later worked for 5 years at Argenta in Harlow before returning to Manchester.
Once again we return to Madingley Hall, Madingley, Cambridge, CB23 8AQ, UK.
A limited number of bedrooms have been reserved for those attending the meeting at the rate of £68 including breakfast. Accommodation should be booked directly with Madingley Hall; contact details and the reference to quote will be provided in the email confirmation upon registration.