Boosting success rates in Pharma R&D through the predictive power of AI-enhanced molecular modelling

BioSimulytics has developed breakthrough technology combining quantum physics, computational chemistry, machine learning and high performance computing to accurately and rapidly predict the most desirable force field energy configuration and crystal structure packing of new drug compounds, thereby supporting the pharma industry in addressing problems with polymorphism and getting from molecules to medicine (M2M) faster and with much greater probability of success.

Download company overview

Thank you

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.

The Need for Innovation in New Drug Development

Global pharma R&D spend is running at >US$200bn with <1% success rate. The industry has seen a doubling of costs to develop a new drug since 2010; at the same time, ROI has fallen from 10% to 2%. Meanwhile demand continues to expand for increasingly personalised medicine to cure devastating chronic diseases. New digital technologies, in particular the use of Artificial Intelligence (AI), are now being embraced to help transform the efficiency and effectiveness of new drug development. Our niche area of focus is on the force-field energy configuration and crystal structure packing of new drug molecules which can manifest in different polymorphic forms.

For API optimization, it is essential to identify and select the most therapeutically desirable and thermodynamically stable polymorph for product development.
Polymorph characterisation is now also essential for compliance with the requirements for regulatory filings on new drugs, as well as for the patenting of new molecular entities (NMEs).
Existing state-of-the-art techniques for polymorph analysis require long and painstaking experimentation by material scientists with uncertain results.
Current computational solutions are also limited because they are typically based on general-purpose force-fields, which have often neither been tested nor optimized to reproduce the various polymorphic forms, including stability ranking, through experiment.

Introducing the BioSim M2M Platform

Shape cubes

Universal solution for any molecule

Our technology has been developed from first principles as a universal solution for any molecule, handling small to very large numbers of atoms, on a linear-scaling basis, factoring in temperature and pressure as well as system-sizing implications on polymorphic forms.

Shape cubes

Ultrafast high performance

Our solution has been built from day one for speed. The core algorithms simplify the incredible complexity of molecular energy configuration and crystal structure packing into actionable real-world outputs, utilising neural network technology to train and fit the models with high precision, that can be processed at high speed.

Shape cubes

Cost effectiveness of the cloud

Our solution runs on AWS High Performance Computing (HPC), delivering all of the cost, convenience and performance benefits of the cloud, with localised data hosting options. Alternatively we also offer an on-premise service utilising the HPC capabilities of our university partners.

Shape cubes

Intelligence to match real-world experimentation

Our solution utilises smart methods to evaluate the thermodynamic stability of crystalline drugs and protein-ligand complexes, determining the free energies at the conditions of interest for drug development, including unique capabilities to handle hydrates.

BioSim M2M Applications

The BioSim M2M Platform is currently being applied to two particularly key areas of R&D in the pharma industry: Crystal Structure Prediction (M2M-CSP) and Structure-based Drug Design (M2M-SBDD).

Crystal structure prediction

The BioSim M2M-CSP solution promises to bring far greater levels of speed and certainty to the current tedious and painstaking work of material scientists in developing new drugs which are fit for patient use, by searching exhaustively and ranking the stability of all crystal polymorphs of a drug compound in its solid-state form.

Structure-based drug design

The universal ability of our technology to accurately compute molecular force-field interactions can be deployed in drug discovery via the M2M-SBDD solution whereby the structure of the target disease protein is known and the binding of a library of drug compounds needs to be determined to identify promising leads with much more predictable success.

Image strip image

Breakthrough Innovation

BioSimulytics has developed a linear-scaling first-principles method for calculating the configurational energy of a system, in particular the configurational energy of a system having a number of particles. The BioSimulytics invention, which is subject to an international patent application, is being applied to crystal structure prediction (CSP) for determining the most stable crystal structure or polymorph of a drug compound, as well as the most stable binding poses in protein-ligand complexes in the arena of structure-based drug design (SBDD) and computer-aided drug design (CADD). A description of some of the core underlying principles behind our breakthrough method as applied to Water-Ice molecules has been published in the Journal of Chemical Theory & Computation.

Download publication

About Us

BioSimulytics was formed in November 2019 as a spin-out company from the School of Chemical and Bioprocess Engineering at University College Dublin (UCD), Ireland. The company is backed by a combination of national funding through Enterprise Ireland and a number of private investors, and has received numerous awards for its innovative technology. BioSimulytics is now working with a number of leading global pharma companies, solution providers and research institutes on implementing its technology to drive major value enhancements in the drug research & development process.

Awards

  • icheme_logo
  • nova_ucd_logo
  • invest_horizon_logo

Meet the Team

Peter F Doyle

Peter F Doyle

CEO & Co-Founder

Peter is a highly experienced co-founder, executive and business coach in the high-tech sector including previous IPO and private equity fundraising experience.

Niall English

Niall English

CTO & Co-Founder

Niall is a Full Professor in Chemical Engineering with previous industrial experience working in molecular simulation and drug design in both the US and UK.

Christian Burnham

Christian Burnham

Head of R&D & Co-Founder

Christian is a physics graduate from Imperial College in London and the lead developer of the company's core technology.

Mozhdeh Shiranirad

Mozhdeh Shiranirad

Computational Scientist

Mozhdeh is a mathematician whose research interests are specialised in the field of quantum computation applied to molecular simulation.

Phúc H. Le Khac

Phúc H. Le Khac

Machine Learning Engineer

Phúc specialises in theoretical to practical aspects of building intelligent systems, with a focus on the application of AI technologies.

Alan Cueva Mora

Alan Cueva Mora

Software Developer

Alan is an experienced software developer with particular interest in Data Science and previous roles working within Oracle and a large financial institution.

Mohammad Reza Ghaani

Mohammad Reza Ghaani

Materials Scientist

Mohammad is an experienced researcher in bioprocess-engineering systems including industrial experience in pharma.

Join our growing team

We are continuously looking for talented scientists and innovators with a passion for making a real impact in the world to join our team.

Enquire now