Join us for this free to attend one-day symposium highlighting new approaches and technologies being applied to the search for future therapeutics.
The day is aimed at scientists from biotechnology, pharmaceutical organisations, not-for-profits and academia who wish to get a broad appreciation of the latest advances in drug discovery delivered by key scientists and thought leaders from leading organisations.
We are delighted to announce a number of high-profile scientists have already confirmed to present at this event including:
Registration is free of charge and only a limited number of seats are available.register
Prof. Gerhard Klebe (University of Marburg)
Keynote: How Predictive are Our Current Methods in Computational Drug Design?
Over the last decades, computational chemists have developed a wide range of methodologies and tools with the ultimate goal of predict binding properties, mostly binding affinity of small molecules towards their target proteins. As a matter of fact, many of these methods are approximate, mainly to maintain their computational tractability, but sometimes also due to our poor understanding of the complexity involved, starting from false or inappropriate simplifications or because we believe that higher accuracy of our considerations is not really required.
To improve our methods and tools we should reflect and validate popular assumptions typically applied in computational approaches. Validations have to be based on experimental evidence or in some cases highly sophisticated computational studies have to be consulted.
In this presentation we will explore some of the limitations in our current methods. These will include the handling of protonation states, consideration of water in ligand binding and misinterpretation of the additivity of functional groups on molecular interactions.
Prof. Jean-Louis Reymond (University of Bern)
Enumerating and Visualizing Chemical Space
Chemical space describes the ensemble of all molecules that are possible by assembling atoms through covalent bonds. In our research we design cheminformatics methods for enumerating and visualizing the chemical space of small organic molecules of interest for medicinal chemistry. Recent databases include the chemical universe database GDB-17 listing 166 billion possible molecules up to 17 atoms of C, N, O, S, and halogens, [1, 2] the fragment database FDB17,  and the ring systems database GDB4c . To understand these databases we use various 3D-visualization tools [5, 6] and innovative fingerprint approaches suitable for big data settings , and use these methods to guide drug discovery projects.
Dr. Marcus Gastreich (BioSolveIT)
Paradigm Changes in Chemical Space Navigation
Pharmacophore-driven navigation through billion-sized virtual chemical spaces are now at a chemist's fingertips. Progress in easy-to-use computational chemistry propels applications that readily deliver tangible or even commercially available molecules, for example as nearest neighbours from giant virtual chemical spaces. One key ingredient for this progress is to include carefully curated knowledge about synthetic procedures in the technology base, keeping the numbers of results that are accessible manageable.
These vast molecular, computable resources enable medicinal chemists to perform rapid scaffold hopping, hit expansion, and structure-activity relationship exploitation in largely IP-free territory and with unparalleled low cost.
The talk will provide an overview of existing methods, outline pros and cons, and report on several worked examples from industry and academia.
Dr. Didier Roche (Edelris)
SBDD From a Diversified NP-inspired Chemical Space
Despite considerable investments in drug discovery, the compound collections used for Hit generation tackle a very small proportion of the 1,030 estimated size of the drug-like chemical space . Although we have observed in recent years the re-emergence of combinatorial approaches to explore considerably larger chemical spaces, notably with DNA encoded libraries (DEL) , annual reviews on drugs entering the market  suggest that the chemical diversity of drugs can hardly be matched with synthetic compound collections. Therefore, making use of structural information, when available, to bias compound design appears to be a must-do strategy.
Since its inception 15 years ago, Edelris has designed and explored a tangible and bio-relevant area of the chemical space, the Keymical Space™, comprising innovative synthetic NP-like, sp3-rich topologies.
This unbiased tangible space has proved to be very attractive for the rational design of innovative target modulators. We will share in this presentation our success stories as well as the limitations encountered in applying virtual screening towards hit generation.
Dr. Edmund Champness (Optibrium)
Imputation of Protein Activity Data Using Deep Learning
The knowledge of compound bioactivity data against drug targets underpins the discovery of new drugs. However, databases are currently sparse; for example, the ChEMBL dataset is just 0.05% compete and the sparsity of data in proprietary pharma databases is similar. We will describe a novel deep learning algorithm to capture correlations within protein activity data, as well as between molecular descriptors and protein activities, to impute the missing activities. Unlike many deep learning methods, this approach is capable of being trained using sparse and variable data, typical of those available in drug discovery. We will present examples illustrating the application of these deep learning networks to impute missing activities in the sparse input data, as well as to make predictions for new compounds based on molecular descriptors alone. The results will be compared with conventional machine learning methods such as random forests and Gaussian processes.
Dr. Andrew Teasdale (AstraZeneca)
Visualisation of PMI-related Risk and its Role in Addressing Regulatory Concerns:
Valsartan — a Case Study
In July 2018 batches of Valsartan were reported to be contaminated with a known carcinogen – Nitroso-dimethylamine (NDMA). Contamination was later found not only in Valsartan but also other Sartans. This presentation will look into the root causes of this contamination, and that of a related Nitrosamine (NDEA), it will examine how the principles outlined in ICH M7 – Mutagenic Impurities, can be applied to investigate risk, principally the use of purge calculations. It will also examine the wider implications of this incident and its possible impact on ICH M7.
Dr. Thierry Hanser (Lhasa Limited)
Privacy-preserving Knowledge Transfer from Corporate Data to Federative Models
Recent progress in the field of artificial intelligence (AI) has dramatically amplified the potential of applying machine learning to many important tasks in the process of drug discovery. To maximise the value of AI applications it is critical to access enough good quality data to allow machine learning algorithms to extract relevant knowledge and produce useful and predictive models. One of the main challenges in AI is therefore to compile such valuable datasets and this task is particularly difficult in the domain of drug discovery due to the confidential nature of the primary information: the chemical structure. Although we have access to limited public data, the most valuable knowledge is embedded in corporate data which cannot be shared easily without disclosing private information. As a consequence, valuable information is kept isolated in private silos for practical reasons despite the willingness of industry to share non-competitive knowledge. To overcome this obstacle, Lhasa Limited has developed a methodology to facilitate the transfer of knowledge from corporate data to federative models whilst preserving the privacy of the original data. The method is based on the Teacher-Student approach  adapted to the domain of molecular informatics. In this presentation, we will show how this methodology can be successfully applied to transfer knowledge from confidential hERG data from pharmaceutical companies into a useful and accurate model without disclosing any chemical structures.
Dr. Christian Kramer (Roche)
Are You Additive? SAR Approaches for Small Molecule Drug Discovery
Additivity is one of the fundamental principles used in medicinal chemistry. It stems from the assumption that a molecule's global property results from the sum of the properties of its individual and independent parts. Additivity is very intuitive for medicinal chemists and is commonly used for SAR analysis and molecular design. In this talk, we will provide a description of several tools that exploit the additivity assumption and aim at helping drug discovery teams understanding their data both in an automated and scalable way. Matched Molecular Pair Analysis together with large chemistry databases can be used to estimate the PhysChem and ADME-Tox effects of substitutions in a data-driven rather than model-driven way. Free-Wilson analysis leads to a direct representation of MedChem SAR knowledge and properly automated can be used to quickly analyse SAR and suggest novel combinations and new promising substituents in a reasonable way. Non-additivity Analysis helps to identify assay errors and cases where the additivity assumption breaks, pointing out cases of complex key SAR that needs to be understood and utilised for optimal design.
SAR analysis tools are an efficient means of organizing data: More time can be spent on using the knowledge generated in better informed design hypotheses. When properly implemented and deployed in computational MedChem tools, additivity-based SAR approaches can have a big impact on the analysis, design, prioritization and experimental procedures applied in drug discovery programmes.
Dr. Rosalind Sankey (Elsevier)
Helping Medicinal Chemists Identify New Opportunities During Lead Identification and Optimisation — Turning High-quality Data into Actionable Insights
With the ever-growing size and number of available datasets, coming from both internal and external sources, the challenge of uncovering key insights and opportunities from this data also becomes increasingly difficult. This presentation will explore a number of approaches and strategies to overcome some of these challenges, including predictive retrosynthesis, matched molecular pair analysis and predictive pharmacology.