Welcome to the Optibrium Community





Forgot login?
Register

Publications & Presentations

Search


html/modules.php!!

Latest Publications & Presentations


Hydrogen Bonding: Ab Initio Accuracy From Fast Interatomic Gaussian Approximation Potentials

Thursday, 23 August 2018 12:26

Mario Öeren gave this presentation at the ACS Fall 2018 National Meeting & Exposition held in Boston, USA.

Abstract

Non-covalent, electrostatic interactions play a significant role in many chemical applications and evaluating their strength is crucial for progress in fields such as drug design and material science. In most cases, due to the nature of these interactions, ab initio calculations are required to accurately assess their strength. However, due to their computational cost, ab initio methods are not suitable for screening datasets with large numbers of structures.

We will present a method based on Gaussian approximation potentials (GAPs), which are interatomic potentials trained on ab initio data using machine learning. While GAPs could be applied to any interaction, we chose hydrogen bonds as an example for this presentation. We will describe the workflow to prepare the GAP training set, how to generate GAPs from density functional theory data using the software QUIP and how to calculate the hydrogen bond energies for a structure from the resulting model. Such an approach allows us to achieve results close to ab initio accuracy, but with significantly lower computational costs. The results are validated against the ab initio calculations and quantum theory of atoms in molecules results.

While GAPs have been mostly used for molecular dynamics simulations of bulk crystals, they can be applied to a variety of problems which require the exploration of a complex potential energy surface (PES); for example, the hydrogen bond energy model described herein can be used in scoring functions for protein-ligand interactions.

Read more...


Robotic Drug Discovery: An Automated Design and Synthesis System to Boost SAR Investigations

Tuesday, 19 June 2018 17:20

Dr Tsukasa Ishihara, National Institute of Advanced Industrial Science and Technology (AIST), gave this presentation at the "Streamlining Drug Discovery" symposium held in Tokyo, Japan on 5 June 2018.

Abstract
We propose an innovative automated architecture to accelerate drug discovery. The system consists of computational drug design programs integrated with robotic compound synthesis apparatus. The computational programs design potentially novel candidates based on tacit knowledge which is automatically extracted from tens of thousands of papers in the medicinal chemistry field, and predict their profiles based on the state-of-the-art machine learning technologies including deep learning. Flow reactors are a key operation device integrated with preparative chromatography to synthesize a series of analogous molecules. Our system has elucidated novel potent compounds comparable to a clinical candidate.

You can download this presentation as a PDF.

Read more...


Library Design for Collaborative Drug Discovery: Expanding Druggable Chemogenomic Space

Tuesday, 19 June 2018 17:17

Dr Kazuyoshi Ikeda, Keio University, gave this presentation at the "Streamlining Drug Discovery" symposium held in Tokyo, Japan on 5 June 2018.

Abstract
Drug discovery data has dramatically increased in the past 20 years due to chemical genomics and automation of screening methods. As a result, we know that chemical space is still immense, and thus it is important to find lead compounds efficiently in the early phases by utilizing information from past drug discovery projects. Recently, collaborative approaches using in silico methods have been successful in expanding the druggable chemical space of screening libraries and identifying hit/lead compounds. Our group (Keio Univ) is constructing an informatics system for analyzing diversity of the compound library collected from a drug discovery screening consortium in Japan. During the talk I will outline examples of how we utilise recent informatics technologies to efficiently design chemical libraries and identify targets.

You can download this presentation as a PDF.

Read more...


A Practical View of Structure Activity Relationship (SAR) Analysis in Novartis Shanghai

Thursday, 14 June 2018 09:43

Dr Zhengtian Yu and Dr Sean Xiao, Novartis, gave this presentation at the "Streamlining Drug Discovery" symposium held in Shanghai, China on 31 May 2018.

Abstract
Structure Activity Relationship (SAR) is the relationship between a chemical structure and its biological activities. SAR analysis is widely used in different stage of drug discovery for desired pharmacological and therapeutic activities. This talk will introduce various tools we use for SAR analysis in CNIBR (Novartis Shanghai), and their using scenarios, as well as an example of a patent analysis workflow, during which use of StarDrop is discussed.

You can download this presentation as a PDF.

Read more...


In Silico Approaches in Genetic Toxicology: Progress and Future

Thursday, 14 June 2018 09:35

Dr Masamitsu Honma, National Institute of Health Science, gave this presentation at the "Streamlining Drug Discovery" symposium held in Tokyo, Japan on 5 June 2018.

Abstract
Currently, more than 130 million chemical substances have been registered in the CAS registry, and this number is increasing at a rate of 4,000/day. Among these chemicals, approximately 120,000 are industrially produced and exist in our environment. To cover the huge number of chemical substances that affect human health, effective screening tools are required. In silico and quantitative structure–activity relationship (QSAR) models defining toxicological endpoints are desirable for regulatory authorities to identify chemicals causing adverse effects without conducting actual toxicological studies.

There has been considerable effort in the development of QSAR models to predict mutagenicity among many toxicological endpoints because mutagenic chemicals pose the highest concern for human health. The recently developed ICH-M7 guideline (Assessment and control of DNA-reactive impurities in pharmaceuticals to limit potential carcinogenic risk) allows the use of the in silico approach to predict Ames mutagenicity for initially assessing impurities in pharmaceuticals. This is the first international guideline addressing the use of QSAR models in lieu of an actual toxicological study for human health assessment. QSAR models for the Ames assay now require higher prediction power to definitely capture mutagenic chemicals. To increase the prediction power, experimental data sets required to build the models are important. A large number of highly reliable data sets are essential to allow the development and improvement of QSAR models. DGM/NIHS in Japan has the largest Ames mutagenicity database, containing approximately 12,000 new chemicals that have not been previously used for developing QSAR models. We provided the Ames data to vendors to improve their QSAR models. The Ames/QSAR international collaborative project, together with 12 QSAR vendors, started in 2014 and has recently been completed. All QSAR models have considerably improved. Some QSAR models showed nearly 90% prediction power, which is the same level as that of the inter-laboratory correspondence of the Ames assay. Using the in cerebro (expert judgement) approach, we can further predict the relevance of mutagenicity data in humans. We are approaching a new era wherein “in silico/in cerebro” will replace “in vitro/in vivo” in genetic toxicology.

You can download this presentation as a PDF.

Read more...


Theoretical Studies of G-Protein-Coupled Receptors

Thursday, 14 June 2018 09:29

Dr Xianqiang Sun, Guangzhou Medical University/ Wuxi Apptec, gave this presentation at the "Streamlining Drug Discovery" symposium held in Shanghai, China on 31 May 2018.

Abstract
The family of G-protein-coupled receptors (GPCRs) contains the largest number of drug targets in the human body, with more than a quarter of the clinically used drugs targeting them. Because of the important roles GPCRs play in the human body, the mechanisms of activation of GPCRs or ligands binding to GPCRs have captivated a great deal of research interest since their discovery. A number of GPCR crystal structures determined in recent years have provided us with unprecedented opportunities to investigate how GPCRs function through the conformational changes regulated by their ligands. This has motivated me to perform molecular dynamics (MD) simulations in combination with a variety of other modeling methods to study the activation of some GPCRs and their ligand selectivity. To address these two issues, we are using opioid receptor, β2 Adrenergic receptor and corticotropin-releasing factor receptors as example to address these two issues and interesting results will be listed in my presentation.

You can download this presentation as a PDF.

Read more...


Genetic Toxicology: Progress on International Test Guidelines and New Methods

Thursday, 14 June 2018 09:24

Dr Yan Chang, National Shanghai Center for New Drug Safety Evaluation and Research, gave this presentation at the "Streamlining Drug Discovery" symposium held in Shanghai, China on 31 May 2018 .

Abstract
Genotoxicity tests are conducted to determine if a chemical or physical agent has the potential to cause mutations or chromosomal damage which, in turn, may lead to adverse health consequences, including cancer, reproductive impairment, developmental anomalies, or genetic diseases. The International Conference on Harmonisation (ICH) S2(R1) guideline (2011) recommends two standard regulatory test batteries, generally includes an assessment of genotoxicity in bacterial and/or mammalian cells in vitro together with rodent assays for chromosomal and/or DNA damage. Several test guidelines of the Organization for Economic Co-operation and Development (OECD) for evaluating the genotoxicity of chemicals and pharmaceuticals have been updated in 2016. The China FDA finalized the revision of the guidance on Genotoxicity Testing and Data Interpretation for Pharmaceuticals this March.

In addition to the current test battery, it was also essential to focus on those technologies most likely to provide an adjunct to, or advantage over, current methods used to predict in vivo genotoxicity and/or carcinogenicity activity to improve human risk assessment. In vivo liver micronucleus test, Pig-a gene mutation assay and humanized in vitro genotoxicity test systems are being developed or in process of inter-lab validation.

You can download this presentation as a PDF.

Read more...


Capturing and Applying Knowledge to Guide Compound Optimisation

Thursday, 14 June 2018 09:17

Ed Champness, gave this presentation at the "Streamlining Drug Discovery" symposia held in Shanghai, China on 31 May 2018 and Tokyo, Japan on 5 June 2018.

Abstract
Compound design requires a combination of knowledge and expertise from different perspectives: Understanding of structure-activity relationships (SAR), based on data from previously studied compounds; expertise from diverse fields to define the multi-parameter optimisation (MPO) objectives of a project; and knowledge of synthetic strategies that may be applicable to create the next rounds of compounds for investigation. All of these forms of knowledge can be captured and applied computationally: Machine learning methods can generate quantitative structure-activity relationship (QSAR) models to predict the properties of novel, virtual compounds; MPO methods capture the desired property criteria for a successful compound for a specific project and rigorously prioritise ideas for consideration; and, optimisation strategies can be captured as structural transformations that reflect steps made in previous chemistry projects.

In this presentation, we will describe these methods and illustrate how they can be seamlessly combined to rigorously explore new, relevant compound ideas and prioritise those most likely to achieve a project objective. This approach can help to stimulate the search for new optimisation strategies and explore a much broader range of compounds than could be achieved based on a single chemist’s or even a project team’s experience. Example applications include the optimisation of compounds with a desired polypharmacology or selectivity profile and exploration of lead hopping strategies to overcome pharmacokinetic issues, while maintaining target potency.

You can download this presentation as a PDF.

Read more...


ICH M7 – Best Practise in Assessing the Mutagenic Potential of Impurities Using In Silico Methodologies

Thursday, 14 June 2018 09:13

Scott McDonald, Lhasa Limited, gave this presentation at the "Streamlining Drug Discovery" symposia held in Shanghai, China on 31 May 2018 and Tokyo, Japan on 5 June 2018.

Abstract
Expert assessment is a fundamental part of the assessment of the mutagenic potential of impurities under the ICH M7 guideline and it makes a specific provision for the application of expert knowledge to support or overturn an in silico prediction. Whilst expert assessment has been successfully applied in this context, there remains some uncertainty as to what constitutes expert analysis.

This presentation will outline a framework to more clearly understand the process of expert assessment under ICH M7 and the role that both in silico prediction methodologies and supporting data play. The use of such a framework can assist in dealing with relatively straightforward situations where the two required in silico methodologies agree to more complicated cases of model disagreement or “indeterminate” or “out-of-domain” predictions.

You can download this presentation as a PDF.

Read more...


Three Key Factors for Success in Molecular Design: Fast, Visual, Easy

Thursday, 14 June 2018 09:06

Dr Marcus Gastreich, BioSolveIT, gave this presentation at the "Streamlining Drug Discovery" symposia held in Shanghai, China on 31 May 2018 and Tokyo, Japan on 5 June 2018.

Abstract
Abou-Gharbia and Childers's 2014 J. Med Chem paper highlighted that the field of therapeutic drug discovery is rapidly changing. According to them, for example more than 1,300 mergers and acquisitions have led to enormous restructurings and widespread job losses. Additional factors such as block buster drug patent expirations, changing educational priorities, and a younger generation of researchers that have grown up with smartphones, have transformed the discipline. This new era exerts an enormous pressure on research and the necessity to adapt software. Such changes, however, must not happen at the expense of the depth of understanding and quality of our tools.

Read more...


Poster: Intuitive Workflow to Enumerate and Explore Large Virtual Libraries

Friday, 20 April 2018 13:13

This poster by Matthew D Segall, Aishling Cooke, James Chisholm, Edmund Champness, Peter Hunt and Tamsin Mansley was presented at the ACS National Spring Meeting 2018 in New Orleans.

Abstract

Enumeration of a virtual library based on cores or scaffolds of interest helps to quickly explore potential substituents around hit or lead series and prioritise strategies that are most likely to yield high quality compounds. In this poster, we will describe a seamless workflow, beginning with a search of commercially available building blocks. These can then be ‘clipped’ to generate the corresponding R-groups for enumeration of virtual libraries, using a flexible and visual approach based on defining substitution points around a substructure search of the building blocks. This flexibility means that chemists are not restricted to a limited number of pre-defined patterns for reagent clipping and can adapt to many different reaction schemes, while the visual interface makes it intuitive and easy to use.

R-group clipping

Read more...


Capturing and applying knowledge to guide compound optimisation

Friday, 20 April 2018 12:38

Matt Segall gave this presentation at the ACS National Spring Meeting 2018 in New Orleans.

Abstract

Compound design requires a combination of knowledge and expertise from different perspectives: understanding of structure-activity relationships (SAR), based on data from previously studied compounds; expertise from diverse fields to define the multi-parameter optimisation (MPO) objectives of a project; and knowledge of synthetic strategies that may be applicable to create the next rounds of compounds for investigation. All of these forms of knowledge can be captured and applied computationally: Machine learning methods can generate quantitative structure-activity relationship (QSAR) models to predict the properties of novel, virtual compounds; MPO methods capture the desired property criteria for a successful compound for a specific project and rigorously prioritise ideas for consideration; and, optimisation strategies can be captured as structural transformations that reflect steps made in previous chemistry projects.

Activity Landscape

Read more...


Translating Methods from Pharma to Fragrances and Flavours

Friday, 20 April 2018 12:25

Tamsin Mansley gave this presentation at the ACS National Spring Meeting 2018 in New Orleans.

Abstract

The pharma sector has generated a wealth of experience in cheminformatics methods that are used in the optimisation of small, ‘drug like’ molecules. While there are differences in the chemistries used to develop flavors and fragrances and the optimisation objectives of these projects, many computational methods can be translated from pharma to guide the design and selection of compounds in this context and improve efficiency and productivity. The properties that describe molecules in these fields are typically different, but both disciplines have the goal of quickly targeting compounds with a balance of properties for the project’s objectives.

In the presentation Tamsin discusses approaches to compound selection and design, including chemical space analysis, property prediction and multi-parameter optimisation, comparing and contrasting datasets and models from pharma with those in flavors and fragrances. This is illustrated by case studies to build and apply robust QSAR models predicting relevant properties, design and prioritisation of new compound ideas and analysis of chemical spaces for selection of compounds, using fragrances and flavors datasets.

Read more...


Pistoia Alliance AI/Deep Learning Projects and Community

Wednesday, 07 March 2018 09:19

At our Drug Discovery Consultants' Day in March 2018, Nick Lynch gave an overview of the Pistoia Alliances' projects and community on AI and Deep Learning, including discussions around best practices and data quality.

You can download his slides as a PDF

Read more...


Imputation of Protein Activity Data using Deep Learning

Wednesday, 07 March 2018 09:07

At Optibrium's 2018 Drug Discovery Consultants' Day, Dr Gareth Conduit from University of Cambridge and Intellegens Ltd. described their deep learning methods for predicting compound activities against protein targets based on sparse training data and presented early results of a collaboration with Optibrium.

You can download his slides as a PDF.

Read more...


Deep Learning and Chemistry

Wednesday, 07 March 2018 08:58

At our 2018 Drug Discovery Consultants' Day, Professor Bobby Glen of the University of Cambridge gave an excellent overview of developments in deep learning and its application to chemistry.

Statistics, Machine Learning and Deep Learning

You can download his slides as a PDF.

Read more...


Preprint: WhichP450

Tuesday, 13 February 2018 16:26

This paper appears in J. Comput.-Aided Mol. Des. and describes the underlying methods and validation of the new model predicting the most likely Cytochrome P450 isoforms responsible for metabolism of a compound in StarDrop's P450 module.

WhichP450 and regioselectivity prediction

Read more...


Discovery Decisions - Collaborating in Data Management

Monday, 22 January 2018 11:58

This paper appeared in the Winter 2018 edition of EBR.

Abstract

From initial hit to development candidate, drug discovery is an iterative process. At each stage, the latest results are reviewed in the context of all the project data, to choose compounds for progression or identify key structure-activity relationships (SAR) that guide the design of new compounds for synthesis. These activities are usually supported by software for data analysis, visualisation and predictive modelling.

However, obstacles remain to the effective use of such software: different applications are often used for each function; scientists may use one to retrieve data from their database, another to visualise their results and a third for predicting properties of new compounds they are considering for synthesis. Just moving and reformatting data for each software application can be time consuming and error-prone. Furthermore, scientists need to learn multiple user interfaces, each with a different ‘look and feel’. Some software, for example visualising protein-ligand interactions in 3 dimensions, may be available only to expert computational chemists, leading to delays while waiting for an expert to be available and the potential for important details to be ‘lost in translation’.

In this article, we will discuss the requirements for a platform to overcome these challenges and support effective decision-making from data to design. Bringing together all of the information revealed by different analyses may reveal new insights and will foster collaboration between different disciplines, leading to more rapid progress and higher quality compounds.

Read more...


Driving Discovery - Predicting P450 Metabolism

Monday, 22 January 2018 11:45

This paper appeared in the Autumn 2017 edition of EBR.

Abstract

Cytochrome P450 (P450) enzymes are responsible for almost 80% of drug metabolism in humans, and metabolism by P450s may lead to several issues for potential new drugs including: low bioavailability, rapid clearance, drug-drug interactions leading to toxicity or lack of efficacy, bioactivation to form reactive or toxic metabolites and variable metabolism in the patient population due to genetic polymorphisms.

In this article, we will discuss some of the questions that a drug discovery team may wish to ask in order to address or, ideally, avoid these issues. For example: Is a compound a substrate for a P450 enzyme and, if so, which isoform? For a compound that is metabolized by a P450, what sites are vulnerable to metabolism, what metabolites will be formed and what strategies could be explored to reduce the rate of metabolism?

In vitro experiments using liver microsomes or hepatocytes can be used address these questions, although more detailed studies are time consuming and expensive. Therefore, computational, or in silico, predictions can be used to supplement experimental data or prioritise compounds for more detailed studies. Furthermore, in silico methods can help to guide the design of new compounds to overcome issues, exploring many optimisation strategies before the medicinal chemist chooses which compounds to synthesise and test. We will describe the state of the art of computational approaches for predicting P450 metabolism and identify areas for future development.

Read more...


Poster: Improved quantum mechanical model of P450-mediated aromatic oxidation

Wednesday, 25 October 2017 13:35

Nick Foster presented this poster at the 21st North America ISSX Meeting 2017 held in Providence, USA.

Abstract

The Cytochrome P450 enzymes (P450s) are a large family of monooxygenases involved in the metabolism of drugs via oxidative reactions such as C-H bond hydroxylation, epoxidation and heteroatom oxidation. It has become increasingly important, within drug development, to develop computer based methods to study and accurately predict P450-mediated metabolism of drugs. We recently published a method that uses quantum mechanical simulations to predict the regioselectivity and lability of cytochrome P450 metabolism . This method uses AM1 calculations and Brønsted relationships to estimate the activation energies for the reaction mechanisms leading to P450 metabolism. This model provides accurate predictions of the regioselectivity of metabolism with faster calculation time than ab initio DFT calculations. However, we have continued to investigate opportunities to further improve the accuracy of the semi empirical methods for some oxidative mechanisms such as aromatic oxidation. In the present study, we model the transition state in the reaction coordinate prior to the intermediates formed during aromatic and aliphatic hydroxylation . The ab initio DFT level of theory is used to model these reactions for a range of P450 3A4 substrates, for which experimental data on relative reaction rates are available. A transition state search is performed to calculate accurate activation energies that correlate well with the experimental data. Subsequently, semi-empirical QM methods are used in a similar transition state search to establish a relationship to these DFT based energies. A correlation between the energetics of DFT and semi-empirical QM methods has been established and this correlation has, in turn, been used to develop an improved predictive model for aromatic oxidation, that can provide a fast and increasingly accurate prediction for the P450 mediated metabolism of drugs.

(1) Tyzack, J.; Hunt, P.; Segall, M. J. Chem. Inf. Model. 2016, 56, 2180-2193.

Read more...


Latest Forums

Read more >