Sponsored Content by Optibrium Ltd.Reviewed by Maria OsipovaSep 18 2025
Discover how Optibrium is transforming early-stage drug discovery through AI-powered software, generative chemistry, and 3D modelling. In this interview, Matt Segall, CEO at Optibrium, shares insights into the company’s scientific innovations, user-focused design, and the future of computational drug discovery.
Please introduce yourself and tell us about your background and current role.
I’m the co-founder and CEO of Optibrium. My original background is in theoretical physics and computer science, but I moved into life sciences applications during my PhD at Cambridge University, where I studied mechanisms for drug metabolism using quantum mechanical simulations, back in the 1990s.
And then in 2001, I moved into biotech to lead a group developing predictive modelling and decision-analysis methods for drug discovery. This eventually led to Ed Champness and me cofounding Optibrium in 2009. We’ve been delighted to work with our teams ever since, supporting an ever-growing client base of pharma, biotech and other life sciences organizations in their drug and compound discovery objectives. Although my role focuses on management and strategy, I still enjoy working with our research team on the latest scientific innovations!
Can you describe Optibrium’s mission and how your software solutions support the drug discovery process from start to finish?
We develop cutting-edge software and AI solutions for small molecule design, optimization and data analysis. Our mission is to build a comprehensive platform powered by novel technologies that improve decision-making and address two of drug discovery’s biggest challenges: efficiency and productivity.
Our platforms support the complex hit-to-lead and lead optimization processes by enabling teams to extract more insights from their data and apply these to design compounds with a higher chance of success. This facilitates better, informed decision-making that makes the most of available time, money, and resources.
How are researchers using Optibrium’s platforms to improve decision-making in the early stages of compound design and optimization?
Our StarDrop™ software enables scientists to target high-quality compounds for their project’s therapeutic objectives early in the drug discovery process, enabling faster progress from hit to candidate. Furthermore, understanding the structure-activity relationships in their chemistry informs the design of new compounds, backed by a comprehensive range of in silico modelling and generative chemistry capabilities. This enables chemists to test a wide range of optimization strategies, then concentrate on those most likely to yield a strong lead or candidate drug, improving the return on experimental investment.
Cerella™ enables drug discovery organizations to more effectively use early-stage data to accurately predict late-stage, expensive outcomes, such as in vivo PK or phenotypic activity. This identifies the most promising compounds and highlights the measurements that will provide the most value to confidently progress compounds to experimental verification. Again, this maximizes the return on experimental investments.
Image Credit: A9 STUDIO /Shutterstock.com
What are some of the key scientific or technological innovations that underpin your AI-driven drug discovery tools?
To target high-quality compounds, we need to combine experimental and calculated data to assess a compound’s chance of success against multiple factors, including potency, ADME and physicochemical properties, and safety. We have pioneered the field of multi-parameter optimization, publishing and patenting several uniquely effective approaches.
Predicting so many factors calls for a broad mix of modelling approaches. We push the boundaries across methods, ranging from mechanistic models based on quantum mechanics, which we use to predict drug metabolism, through 3D molecular modelling to optimize target binding, to machine-learning methods that use existing data to predict new compounds.
Beyond science, it’s really important that we deliver these platforms conveniently for our customers at a low cost of ownership. We were delighted to partner with Amazon Web Services as one of the first scientific applications for their AppStream platform. This platform enables our highly visual and interactive StarDrop application to be accessed via a browser as a fully hosted SaaS solution as well as a traditional on-premises installation.
How does generative chemistry fit into Optibrium’s broader vision for molecular design, and what impact is it having on real-world research?
Generative chemistry has massive potential in drug discovery. The sheer magnitude of chemical space means scientists simply can't explore it meaningfully alone.
By combining AI and machine learning with generative chemistry, we can probe chemical space faster and more comprehensively. This means we can ensure the best drug candidates are found, and valuable opportunities are not overlooked.
But we do need more than just generative chemistry to make these findings; an expert's scientific knowledge and strategic understanding of the project are essential inputs. We call this combination of human expertise supported by AI algorithms in drug discovery Augmented Chemistry®. The Inspyra™ module in our StarDrop platform has a seamless feedback loop between generative chemistry algorithms and an expert chemist, to guide the algorithms to explore the most relevant chemical space to quickly identify optimal compounds.
Optibrium’s tools are known for balancing rigorous science with usability—how do you approach designing software that works seamlessly for both computational chemists and broader R&D teams?
One of our specialities at Optibrium is making sophisticated computational methods accessible to experimental scientists through a highly intuitive and visual user interface. Results must be presented clearly and in a way that’s easy to interpret, to enable users to make effective decisions quickly. To achieve this, we think carefully about human-computer interaction and work closely with our users to understand how they think about their compounds and data in the context of their project objectives.
It's also important to facilitate close collaboration between computational experts and the chemistry and biology teams they support, and we achieve this in two ways. The new collaboration capabilities coming soon in StarDrop enable project team members to easily share the results of their analyses, helping the whole team to benefit from their insights. We also offer many flexible ways to customize and integrate StarDrop with in-house platforms, enabling computational chemists to incorporate their own models and algorithms, thereby increasing their impact on drug discovery projects by making them easily accessible.
Can you walk us through how Optibrium’s 3D molecular modelling capabilities are helping scientists visualize and optimize compounds more effectively?
Understanding a molecule's three-dimensional structure and interactions is essential to understanding its binding interactions and drug-like properties. When you only think in 2D, you can miss this vital information that guides the design of better candidate drugs.
Our BioPharmics platform provides industry-leading 3D ligand and structure-based design technologies. We acquired BioPharmics in 2023, and our colleagues and co-founders of BioPharmics, Ajay Jain and Ann Cleves, have continued to extend this technology and demonstrate its superior performance through published research and industry collaborations.
Benchmarks conducted by global pharma companies show that BioPharmics is more effective at identifying active compounds during virtual screening, meaning that you have to test fewer compounds experimentally.
Recent research has achieved remarkable accuracy in predicting how compounds fit within protein targets. This insight enables chemists to optimise potency and properties more efficiently, reducing the need for extensive synthesis and testing. The ultimate goal, however, is to predict compound affinity—the ‘holy grail’ of computational chemistry.
The leading technology, free-energy perturbation (FEP), works well in some circumstances, but is limited in that it requires an experimentally-determined protein-ligand structure, it can only predict small changes from a reference ligand and is computationally very expensive, requiring expensive GPUs. The QuanSA method in the BioPharmics platform is equivalent in accuracy, is 1000x faster without GPUs, and can be applied to much larger changes in chemical structure, even to different chemical series.
We’re also pioneering applications to ‘beyond rule-of-five’ compounds such as peptidic macrocycles. Despite their therapeutic potential, the size and flexibility of these molecules pose an enormous challenge that other molecular modelling software can’t tackle. By more quickly and rigorously exploring the vast number of potential conformations of macrocycles and combining this with experimental biophysical data, our technology makes effective 3D modelling of these molecules accessible for the first time, opening the field to the efficiency improvement this already brings to conventional small-molecule drug discovery.
In what ways are your platforms supporting collaboration among multidisciplinary teams in pharmaceutical and biotech organizations?
Collaboration is essential to drug discovery. Bringing together multiple disciplines, including chemistry, biology, and DMPK, accelerates discovery. Plus, increasingly geographically dispersed teams need to work together effectively to make faster progress.
To support effective collaboration, we must ensure that everyone on the team has access to the latest, accurate data and insights, to prevent any wasted time pursuing ideas or conclusions that colleagues have already explored.
Our recent announcement of the upcoming version 8 of our StarDrop platform will embed all its design and analysis capabilities in a real-time collaboration environment. This means that everyone on a project is working with the same, up-to-date compounds and data, while being able to view information in a personalized, meaningful layout to perform their own analysis. Results and decisions are then instantly shared back with colleagues.
As AI becomes increasingly prominent in drug discovery, how do you ensure the robustness, transparency, and scientific integrity of your models?
We have the highest standards across every aspect of our science. Our research team extensively tests and validates all new methods before rolling them out into our software. We also prioritize scientific integrity by continuing to publish in peer-reviewed journals independently and in collaboration with pharmaceutical and biotech companies.
A key aspect of our approach to building AI models is assessing the uncertainty in each prediction. This information can be just as important as the result itself. To use a result confidently in guiding decisions, researchers need to understand just how much they can trust it. Areas of high uncertainty can also reveal new places to explore and find promising opportunities that might have been overlooked.
What are some of the biggest challenges your users face in early-stage drug discovery, and how is Optibrium helping to overcome them?
Time and cost. Drug discovery remains a slow and expensive process. Our software addresses this by enabling teams to extract more information from their data, facilitating better and more informed decision-making. This not only accelerates the discovery process by reducing the number of compounds that need to be synthesized, but also makes it significantly more efficient by avoiding wasted experimental efforts, ruling out those measurements that won't add value before resources are committed.
A lesser known challenge of drug discovery is opportunity cost. Given the complexity and noise in the data we generate, it’s all too easy to incorrectly discard potentially valuable candidate drugs. Our unique approaches can highlight potential missed opportunities caused by lost, uncertain, or incorrect experimental data for further investigation.
We also recognize that some organizations may lack the necessary IT infrastructure to support valuable computational platforms. So we are able to take the support burden off their hands and reduce the cost of software ownership with a cloud-based version of StarDrop that provides the same molecule design, optimization, and data analysis capabilities as our desktop version. We handle all the infrastructure, maintenance, and updates, so our customers can focus entirely on researching and discovering new medicines without worrying about IT overhead.
Looking ahead, what areas of research or software development are most exciting to your team at Optibrium?
We're always looking for new ways to bring value to our customers, whether that's through new science, new technology, or improved user experience. We're working on further developing and enhancing our AI and machine learning methods and our 3D modelling capabilities, but we're also working to make these incredibly powerful methods even more accessible to medicinal chemists whilst continuing to surface insights in clear and interpretable ways for discovery teams.
Next year, we will be starting a new research project that will apply one of the latest areas of machine learning to improve our methods of predicting how compounds will be metabolized in the body. It has the potential to drastically reduce the computational cost of running highly accurate simulations, which would enable teams to run calculations on many more molecules. Once developed, these models will be integrated directly into StarDrop, making them available to biopharma companies worldwide.
Also, as I mentioned earlier, we are moving quickly towards the release of StarDrop 8 at the end of the year, which will transform how teams can use our software to facilitate collaboration.
Where can our readers learn more?
You can find all the information on the features and applications of our software portfolio on our website at www.optibrium.com, or contact us at [email protected] if you want to speak with one of our experts. We also regularly update our Knowledge Base on our website with useful content, including blogs addressing important topics in drug discovery, computational chemistry and machine learning, case studies, publications and posters that show the real-world impact of our technology.
You can also follow us on LinkedIn to stay updated on our news, product launches, and conference attendance.
About Matthew Segall 
Matthew Segall is CEO of Optibrium. He has an MSc in Computation from the University of Oxford and a PhD in theoretical physics from the University of Cambridge. Since 2001, Matthew has led teams developing predictive models and intuitive decision-support and visualization tools for drug discovery. Matt has published over 40 peer-reviewed papers and book chapters on computational chemistry, cheminformatics and drug discovery. In 2009 he led a management buyout of the StarDrop business to found Optibrium, which develops novel technologies and ground-breaking AI software and services, including Cerella and Inspyra, that improve the efficiency and productivity of drug discovery.
About Optibrium Ltd.
Optibrium provides elegant software solutions for small molecule design, optimization and data analysis. Optibrium's portfolio of products includes:
- StarDrop™, which brings confidence to the selection and design of high quality candidate compounds. StarDrop creates an intuitive, highly visual and flexible environment to facilitate and speed up lead identification and optimization, quickly targeting effective candidate compounds with a high probability of success downstream.
- Cerella™, a deployable AI platform that creates new value from drug discovery data, revealing hidden insights into the best compounds and most valuable experiments for each project. Cerella makes confident predictions, accurately filling in missing values, especially for expensive downstream experiments that can’t be predicted by other methods. This enables teams to do more, even with sparse, limited data sets.
- BioPharmics™, a platform for fast, accurate, and robust 3D modelling from small molecules to large macrocycles. With industry-leading ligand- and structure-based design capabilities, BioPharmics quickly generates accurate conformational ensembles, predicts bound ligand poses and binding affinities without requiring protein structural information, and enables high-performance virtual screening at scale.
Founded in 2009, Optibrium continues to develop new products and research novel technologies to improve the efficiency and productivity of the drug discovery process. Optibrium works closely with its broad range of customers and collaborators, including leading global pharma, agrochemical and flavoring companies, biotech and academic groups.
Optibrium introduces cloud-based version of StarDrop drug discovery platform