Welcome to the Optibrium Community

Forgot login?

Publications & Presentations



Latest Publications & Presentations

New UK Collaborative Uses AI to Predict Missing Data Points in Compound Data

Wednesday, 13 March 2019 16:52

A new UK collaboration focuses on taking sparse data – data where a significant amount of points are missing from the complete sets – or “noisy” data – data where a significant amount of variables could contribute to issues and changes in results – and making predictive models that fill in missing points with degrees of certainty and without having to undergo costly experimentation.

You can download the article here as a PDF

You can link to Rx Data here


SBDD From a Diversified NP-Inspired Chemical Space

Wednesday, 13 March 2019 11:44

At the 2019 Streamlining Drug Discovery Symposium in Frankfurt, Didier Roche from Edelris presented 'SBDD From a Diversified NP-Inspired Chemical Space'.

You can download the presentation slides here as a PDF


Turning High Quality Data Into Actionable Insights

Friday, 22 February 2019 14:33

At the 2019 Streamlining Drug Discovery Symposium in Frankfurt, Rosalind Sankey (Elsevier) presented 'Helping Medicinal Chemists Identify New Opportunities during Lead ID and Optimisation - Turning High Quality Data into Actionable Insights'.

You can download the presentation slides here as a PDF


Capturing and Applying Knowledge to Guide Compound Optimisation

Friday, 22 February 2019 13:23

This article was published in the online edition of Drug Discovery Today; February 2019

Matthew Segall, Tamsin Mansley, Peter Hunt, Edmund Champness


Successful drug discovery requires knowledge and experience across many disciplines and no current ‘artificial intelligence’ method can replace expert scientists. However, computers can recall much more information than any individual or team and facilitate transfer of knowledge across disciplines. We’ll discuss how knowledge relating to chemistry and the biological and physicochemical properties required for a successful compound can be captured. Furthermore, we’ll illustrate how, by combining and applying this knowledge computationally, a much broader range of optimisation strategies can be rigorously explored, and the results presented in an intuitive way for consideration by the experts.

You can download the article as a PDF


Are You Additive? SAR Approaches for Small Molecule Drug Discovery

Thursday, 21 February 2019 11:10

At the 2019 Streamlining Drug Discovery Symposium in Frankfurt, Christian Kramer gave this insightful presentation 'Are You Additive? SAR Approaches for Small Molecule Drug Discovery'.

Optibrium Community Members can download this presentation


Imputation of Assay Bioactivity Data using Deep Learning

Thursday, 14 February 2019 10:24

This preprint paper was accepted for publication in the Journal of Chemical Information and Modeling. 13 February 2019


We describe a novel deep learning neural network method and its application to impute assay pIC50 values. Unlike conventional machine learning approaches, this method is trained on sparse bioactivity data as input, typical of that found in public and commercial databases, enabling it to learn directly from correlations between activities measured in different assays.

In two case studies on public domain data sets we show that the neural network method outperforms traditional quantitative structure-activity relationship (QSAR) models and other leading approaches. Furthermore, by focussing on only the most confident predictions the accuracy is increased to R2 > 0:9 using our method, as compared to R2 = 0:44 when reporting all predictions.

You can download this paper as a PDF


AI Advances Healthcare Research

Monday, 10 December 2018 14:23

Harnessing AI for drug discovery applications will significantly speed the identification of promising drug candidates, believes Matt Segall, CEO at Optibrium. The UK-based firm, together with partners Intellegens and Medicines Discovery Catapult, recently received a grant from Innovate UK to help fund a £1 million project focussed on combining Optibrium’s existing StarDrop software for small molecule design, optimisation and data analysis, and Intellegen’s deep learning platform Alchemite.

The aim is to develop a novel, deep learning AI-based method for predicting the ADMET (absorbtion, distribution, metabolism, excretion and toxicity) properties of new drugs candidates. Ultimately, the platform could help to guide the selection and design of more effective, safer compounds earlier in the discovery process...


You can link to the Scientific Computing World article here


Managing Internal and External Chemistry for Efficient Drug Delivery

Friday, 30 November 2018 13:50

At the 2018 Streamlining Drug Discovery Symposium, USA, David Hollinshead (Elixir Software) and Andrew Griffin (Praxis Precision Medicines) presented Managing Internal and External Chemistry for Efficient Drug Delivery.

You can download the slides here as a PDF


Electrostatic Complementarity as a New Approach to Visualize and Predict Activity

Thursday, 29 November 2018 15:54

At the 2018 Streamlining Drug Discovery Symposium, Sylvie Sciammetta presented Electrostatic Complementarity as a New Approach to Visualize and Predict Activity.

You can download her slides here as a PDF


Medicinal Chemist’s Relationship with Additivity

Wednesday, 28 November 2018 15:12

Medicinal Chemist’s Relationship with Additivity: Are we Taking the Fundamentals for Granted?

At the San Francisco, Streamlining Drug Discovery Symposium 2018, J. Guy Breitenbucher from UCSF gave this in-depth presentation.

You can download the presentation slides here as a PDF


Bigfoot, the Loch Ness Monster, and Halogen Bonds

Thursday, 22 November 2018 10:37

At the 2018 Streamlining Drug Discovery Symposium in San Diego, David Lawson treated us to this illuminating presentation entitled Bigfoot, the Loch Ness Monster, and Halogen Bonds: Separating Rumors from Reality in Molecular Design.

You can download his slides here as a PDF


A Novel Scoring Profile for the Design of Antibacterials Active Against Gram-Negative Bacteria

Friday, 16 November 2018 12:15

At the 2nd SCI/RSC Symposium on Antimicrobial Drug Discovery, 12-13 November 2018, Bailey Montefiore, Optibrium - Franca Klingler, BioSolveIT - Nicholas Foster, Optibrium presented 'A Novel Scoring Profile for the Design of Antibacterials Active Against Gram-Negative Bacteria'.


The increasing occurrence of multidrug-resistant bacteria is one of the major global threats to human health. Design of new antibacterials is challenging because new compound classes often do not possess the unique physicochemical properties required to penetrate the gram-negative cell wall. It is accepted that the physicochemical properties of many drugs are similar and attempts have been made to characterise these ‘drug-like’ properties, such as Lipinski’s ‘rule of five’ for orally dosed drugs. However, antibiotics are a known exception to these rules. We compared antibiotics active against gram-negative bacteria with other classes of drug and compounds considered in medicinal chemistry projects to determine criteria for selection of compounds with a higher chance of success as a gram-negative antibacterial. These criteria are based on calculated properties, so can help to guide the design and selection of compounds in discovery projects.

You can download the poster presentation as a PDF


Imputation of Protein Activity Data Using Deep Learning

Wednesday, 14 November 2018 15:44

At the US Symposia, Streamlining Drug Discovery 2018 in Cambridge MA, Matthew Segall from Optibrium and Tom Whitehead from Intellegens presented Imputation of Protein Activity Data Using Deep Learning.

You can download the complete slide presentation as a PDF


WaterSwap to Assess Target Druggability

Monday, 12 November 2018 15:07

At the 2018 Streamlining Drug Discovery Symposium in San Diego and San Francisco, Adam Kallel gave an insightful presentation on WaterSwap to Assess Target Druggability.

You can download his slides as a PDF


Using AI to Improve the Safety of New Drug Candidates

Thursday, 08 November 2018 08:37

On 18 October 2018, at the Streamlining Drug Discovery Symposium in Cambridge MA, Nigel Greene gave this fascinating presentation on Using AI to Improve the Safety of New Drug Candidates.

You can download his slides as a PDF


Two Decades under the Influence of the Rule of Five and the Changing Properties of Approved Oral Drugs

Monday, 29 October 2018 09:06

This paper appeared in Journal of Medicinal Chemistry, September 13, 2018.


Two decades have passed since the rule of five ushered in the concept of “drug-like” properties. Attempts to quantify, correlate, and categorize molecules based on Ro5 parameters evolved into the introduction of efficiency metrics with far reaching consequences in decision making by industry leaders and scientists seeking to discover new medicines. Examination of oral drug parameters approved before and after the original Ro5 analysis demonstrates that some parameters such as clogP and HBD remained constant while the cutoffs for parameters such as molecular weight and HBA have increased substantially over the past 20 years. The time dependent increase in the molecular weight of oral drugs during the past 20 years provides compelling evidence to disprove the hypothesis that molecular weight is a “drug-like” property. This analysis does not validate parameters that have not changed as being “drug-like” but instead calls into question the entire hypothesis that “drug-like” properties exist.


Antimalarial Lead-Optimisation Studies on a 2,6-Imidazopyridine Series within a Constrained Chemical Space To Circumvent Atypical Dose−Response Curves against Multidrug Resistant Parasite Strains

Wednesday, 24 October 2018 07:58

This paper appeared in Journal of Medicinal Chemistry, September 26, 2018.


A lead-optimization program around a 2,6-imidazopyridine scaffold was initiated based on the two early lead compounds, 1 and 2, that were shown to be efficacious in an in vivo humanized Plasmodium falciparum NODscidIL2Rγnull mouse malaria infection model. The observation of atypical dose–response curves when some compounds were tested against multidrug resistant malaria parasite strains guided the optimization process to define a chemical space that led to typical sigmoidal dose–response and complete kill of multidrug resistant parasites. After a structure and property analysis identified such a chemical space, compounds were prepared that displayed suitable activity, ADME, and safety profiles with respect to cytotoxicity and hERG inhibition.


High-Quality Hits from High-Throughput Screens

Monday, 15 October 2018 13:21

This paper appeared in Genetic Engineering & Biotechnology News, October 15, 2018.


When analysing the results from a high throughput screening (HTS) campaign the goal is to identify diverse hit series with high activity, structure-activity relationships (SAR) that indicate the opportunity for further optimisation and good ‘lead like’ properties. The common practise is to apply filters to these large datasets, for example an activity threshold or simple properties such as molecular weight, logP, numbers of hydrogen bond donors and acceptors or the presence of substructures that may indicate non-specific binding. However, this process draws artificially harsh distinctions between compounds, given the inherent variability in HTS data and the low correlation between simple properties and the ultimate in vivo disposition of a compound. This leads to selection of ‘false positives’, i.e. active compounds that are not good starting points for further optimisation and ‘false negatives’, i.e. potentially good compounds that have been inappropriately rejected. We will illustrate how a true multi-parameter approach enables appropriate weight to be given to these data to confidently identify high quality, potent hits while avoiding missed opportunities.

Mapping this information across the chemical diversity of the compounds explored in an HTS campaign, by clustering or visualisation of a ‘chemical space’, helps to find ‘hot spots’ representing high quality series of compounds for further investigation while also considering diverse chemistries to provide potential backup series. Finally, exploring the SAR within these series then helps to identify further opportunities for optimisation. We will show how this can all be achieved in a high visual and intuitive way, to move quickly and confidently from initial HTS hits to high quality lead series.


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.


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.


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.

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.


Latest Forums

Read more >