CSIRO’s Dr Dave Winkler (left) and Dr John Tsanaktsidis (right) using MolSAR technology on a laptop.

Computer aided drug discovery

CSIRO has developed a powerful software modelling tool, MolSAR, which has the potential to reduce drug development costs by up to US$50 million per drug.

• 10 September 2007 | Updated 14 October 2011

• Overview of project
• Current activities
• Outcomes
• About the scientist

Overview of project

Pharmaceutically relevant knowledge-based technologies are revolutionising the modern drug discovery process.

The ability to extract meaningful information from data sets, design new biologically active structures and to predict their properties is essential for competitive drug discovery in the 21st century.

CSIRO has developed a powerful predictive computational tool, MolSAR™, which is capable of generating robust structure-property models for specific types of biological responses for use in drug, veterinary and agrochemical discovery and development.

Using a Bayesian neural network, MolSAR™ is able to identify very complex relationships between chemical structure and biological activity.

Current activities

'MolSAR™ means that a lot of the work to find and develop new drugs can now be done on a computer,' says research leader Dr Dave Winkler.

'MolSAR™ can be used to extract useful information from large quantities of data, for example from pharmacological experiments, which can then be used in the drug discovery process.'

'It can also search through existing collections of chemicals to find new drug leads and help design new chemical collections which have a higher likelihood of containing good drug leads.'

'Another benefit is that it can 'pre-test' a possible chemical, so MolSAR™ could find that the compound may work as a drug against the disease but have unwanted side effects.'

Knowledge-bsed technologies have an increasing role to play in the costly business of discovery and development of new pharmaceuticals.

'If technologies like MolSAR™ could save one year in the drug development process, this would save industry at least US$50 million per drug,' says Dr Winkler.

MolSAR™ can be used for a wide range of problems related to the human pharmaceutical discovery process, as well as other commercially important areas such as agrochemical and veterinary drug discovery.

It can help replace animal testing of drugs, cosmetics and industrial chemicals.

As the mathematics behind MolSAR™ is applicable to a wide range of pattern recognition problems, it could be applied to other technology design systems relating to biomaterials and stem cell biology for example.

Our specialist team can design, synthesize and develop biologically active molecules targeted to customers’ requirements.

Outcomes

The MolSAR™ team have been approached by a major corporation to licence the technology.

The licence agreement is being finalised and the technology transfer is effectively complete.

MolSAR™ should provide CSIRO with significant return on investment.

About the scientist

Dr Dave Winkler has a doctorate from Monash University, Melbourne, Australia, and is:

• a Fellow of the Royal Australian Chemical Institute
• an Adjunct Professor at Monash University
• a member of the Executive of the Asian Federation for Medicinal Chemistry
• a Board member of the international QSAR and Modelling Society.

Find out more about CSIRO’s Health & Wellbeing research.