The Physiome Project

What is the Challenge? (The Need)

"If Boeing developed aircraft the way the pharmaceutical industry develops drugs, they would develop ten very different aircraft, fly them, and the one that stayed in the air would be the one they would sell."
   -- Tom Paterson, Chief Scientific Officer, Entelos, Inc.

When it comes to bioengineering, the "state of the art" is still mainly "hit or miss" and often the mechanism by which the "hits" work are not fully understood. Tom Barnes, the Deputy Vice Chancellor of University of Auckland, points to the world-class work that Professor Peter Hunter is doing in New Zealand. Peter Hunter is the Director of the Bioengineering Research Institute (BRI) of the University of Auckland (http://www.physiome.org.nz). Peter and his team are working on, among other things, ways to dramatically change the process of drug discovery. Under the current process, for every 10,000 compounds that are investigated, only one will result in a new drug.

Working as part of the international Physiome Project, BRI is working hard to bring together vast amounts of existing experimental data and use that to better understand, or model, the entire organism. Their efforts will link the larger structures (like the fluid dynamics of the heart - see figures) to the smaller molecular structures (e.g., cells, proteins, DNA). Such biological systems are vastly more complex than human engineered systems and understanding them will require specially designed software and instrumentation and an unprecedented degree of both international and interdisciplinary collaboration.

To that end, the University of Auckland is working not only on the mathematical models, but also on ways to collect and share data with their international colleagues. Using their quantitative models and data, the effects of new drugs can be modelled in silico (on the computer) thus cutting out years of clinical testing. Eventually, a patient's diagnosis and treatment will be based partly on diagnostic equipment and partly on a complex model of that individual's physiome.

All of this needs computing power, way beyond what one institution can muster on its own.

How Can NGI Help? (The Use)

The bottleneck to achieving the goal of mathematically modelling the human body is two-fold: 1) the requirement of massive raw processing power and 2) the manipulation and storage of vast amounts of data. That's where the Next Generation Internet (NGI) comes in.

On the Processing side, while the University of Auckland has an IBM supercomputer to help with the modelling, this is not enough. Collaborative research such as this is best done by connecting supercomputers into a GRID (it's sometimes likened to the computer-networking equivalent of the electricity grid, offering processing power on demand). BRI can look at doing that, however a GRID needs NGI. For example, if the University of Auckland and NIWA were linked up using NGI, then spare capacity at NIWA (NIWA's supercomputer has 17.6 Terabytes of storage capacity) or even Massey University can be harnessed by BRI. In the absence of an NGI network in New Zealand, this is not economically and indeed technically feasible. We are talking about a new class of networking way beyond the typical 10 Mbps connections to the Internet of today. These links need to start at 1Gbps and move into "Terabit" territory. With such a connection, the Physiome Project hopes to spread the computational load across several supercomputers at multiple sites around the world.

On the Data side, Professor Barnes commented that the connectivity barriers are already becoming a constraint and will become "severe in the future" as the amount of available data continues to grow exponentially while current connections speeds remain in the low Mbps territory.

From a Physiome modelling perspective, the use of NGI is quite literally the only way to unlock the functions and operations of the human body.

The Bottom Line (The Value)

Utilising NGI connections, New Zealand has the capability of becoming a world leader in bioengineering-based modelling. The benefits to the New Zealand economy include:

• Intellectual Property surrounding the actual modelling process itself. This IP is expected to generate significant royalties and contracts (e.g., for drug discovery and any other medical procedure evaluations), providing plenty of work for the New Zealand scientific community.
  
• Intellectual Property resulting from tested compounds. These results could be maintained in a database service (i.e., bioinformatics) and access could be provided internationally through NGI-based high-speed connectivity. Using this service, other research groups could trim years off their research efforts.

What would the above Intellectual Property be worth? "Developing one new drug today takes an estimated 12-15 years and costs over US$500 Million. Over US$150 Million of that is attributed to drug and experiment failures, with a failure rate in clinical trials of over 80%" (www.entelos.com/science/insilico.html).

In 2002, there were around 20 new drugs approved by the US FDA. If modelling could shave 5 years off of development and prevent 1/3 of the money spent on experimental failures, and if companies were willing to spend only 20% of the potential savings to realise them, and if New Zealand captures only 10% of the resulting market share, then this could still produce revenues upwards of NZ$120 million per annum for participants of the sector (such as BRI) with strong future growth possibilities.

This is not science fiction; indeed Professor Hunter is on the advisory board of Physiome Sciences, a US-based company whose mission is to help pharmaceutical companies develop better drugs faster through the use of biological simulations. We can only hope that New Zealand rises to give Professor Hunter a hand by investing in NGI while this opportunity still exists.