Bio-IT down under: supply without demand?

It is a question of making haste slowly for Australia's small circle of private companies dedicated to selling bioinformatics-related goods and services. The good news is that new arrivals are seeping into the sector while established players are consolidating their positions and looking forward optimistically.

The pace seems quicker in the layer of publicly-funded research institutes which largely drive bioinformatics research in Australia. The four leading members of that group are all getting stronger and large blocs of Commonwealth cash have been released to two others who provide services to proteomics and genomics researchers.

But the X factor -- demand for bioinformatics services from the commercial biotech community -- is still largely missing.

Even so, when asked if the sector's health is improving, the answer is a resounding "yes", from Dr Tim Littlejohn, a tireless proselytiser for Australian bioinformatics and founder of a number of commercial companies in the field, the most recent being bioinformatics training company BioLateral which is co-located with the Garvan Institute of Medical Research in Sydney.

"It is small, fragile, and embryonic but it is alive and for a small fragile embryo, well!"

The terms bio-IT and bioinformatics are often used interchangeably. Purists prefer to reserve bioinformatics as an alternative term for computational biology while applying bio-IT more to the engineering that research into hardware and software products for end users. Littlejohn defines bioinformatics as the use, management and exploitation of biological data using information technologies.

Whatever the terminology, the discipline is applied in the biotechnology and pharmaceutical industry, hospital-based research, universities and government scientific research establishments.

Bioinformatics tools and services underpin gene discovery (high-throughput genetic sequencing), gene function (chips and microarrays, proteomics), computer-aided drug design and molecular modelling plus management of clinical trial efficacy data, toxicology studies and patient data investigations.

In terms of research, Australia's bioinformatics space is dominated by a cluster of publicly-funded centres which are generally associated with universities or medical institutes. In the vanguard of this group are the Victorian Bioinformatics Consortium (VBC) at Monash University in Melbourne, the Centre for Bioinformation Science (CBIS) at the Australian National University, and the bioinformatics units of Walter and Eliza Hall Institute (WEHI) in Melbourne and the Institute for Molecular Bioscience (IMB) at the University of Queensland. Bioinformatics also has a growing presence in Tasmania, Western Australia and New South Wales (see breakout, States of play). Set up just over a year ago, the VBC appears to be performing well in the role of a state-funded central organisation set up to service the specialised bioinformatics needs of a variety of both privately and publicly-funded organisations.

Its staff of 12 to 15 provides a critical mass of bioinformatics specialists that wouldn't otherwise exist in Victoria.

The centralist approach works more effectively than distributing the same number of specialists in ones or twos across a variety of independent organisations, says VBC director Prof Ross Coppel.

That scenario too often degenerates into a system of "plugging leaks and crisis management" rather than producing well-orchestrated long-term results, he says.

Besides acting as a bioinformatics brains trust for government agencies and private companies seeking assistance in research projects, the VBC develops new application tools and works with individual biologists to extract meaning from their datasets.

Coppel says there is growing interest among federal agencies in coordinating bioinformatics efforts.

The consortium is registering an "enormous" increase both in its range of activities and the number of projects it has been asked to participate in, he says.

"In terms of meeting a need in the research community for advanced bioinformatics assistance, we are getting a lot of interest which translates into challenging projects."

"Everyone recognises that a vibrant biotech sector, particularly in relation to intellectual property, can't exist without bioinformatics and that if the major bioinformatics requirements aren't met, researchers will fall behind.

"Because that has been accepted, our research activities are booming. We have a lot of different projects underway which we believe will have beneficial outcomes in terms of accelerating IP discovery in areas such as new biological enzymes in important biological processes which could form the basis for downstream products."

Research focus

IN the VBC's peer group of research institutes, the efforts of CBIS centre on phylogeny and genetic mapping , IMB's on genomics and WEHI's on microarrays. The VBC works in some of the same areas, for example in genomic analysis, where it is particularly interested in the prediction of function from sequence information, and in functional genomics analysis involving microarrays.

However the consortium is stepping up its work in the field of integrative databases, and developing techniques for dealing with very complicated databases consisting of large amounts of multidimensional data from various sources which need to be tied together.

The research has practical applications for biology researchers, as is the case with most of the consortium's work, according to Coppel.

"We are quite interested in the issue of database applications which collate all the information that might be available about a particular organism."

"For example, if a researcher wants a current snapshot of all research published on a certain topic, an intelligent agent is needed to scan the literature and extract the right bits to plug into the right parts of the database."

The ability to conduct literature searches of that type feeds into integrative database projects on which the VBC is involved such as the creation of databases of mutation data in genes that cause significant diseases in humans.

The work requires expertise related to syntactic meaning, data mining and other elements that call for partnerships with IT researchers, Coppel says.

The consortium is well-placed to form partnerships with specialists in other spheres because of its structure. Its core members include Monash University's faculties of Medicine, Nursing & Health Sciences and Information Technology, the CSIRO, and Agriculture Victoria's Plant Biotechnology Centre.

The grouping guarantees any project the consortium launches has access to advanced expertise in medicine, biological science, agricultural and veterinary research as well as computer science.

Who's interested?

Biotech companies are under-represented in the VBC's client base and to date the consortium has seen little interest from private sector biotechs in using bioinformatics to underpin the process of IP discovery, says Coppel.

A sprinkling of private sector biotechs have strong in-house bioinformatics capabilities -- one that Coppel points out is drug discovery company Cytopia -- but they form the exception than the rule.

On the other hand, the VBC has strong links with major IT vendors from IBM to Hewlett Packard/Compaq who are vying with one another to provide the hardware and software needed in the bio-IT space.

"We are working with number of commercial IT providers either to develop new solutions for bioinformatics problems or to port IT products to the biotech sphere which have been developed originally for applications outside biology."

Another bright spot for publicly-funded bioinformatics research institutes has been the guarantee of an injection of about $30 million over the next five years into the Australian Proteome Analysis Facility (APAF) and the Australian Genome Research Facility (AGRF).

APAF was awarded $16.5 million and AGRF about $14 million under the Commonwealth Government's Major National Research Facility (MNRF) scheme.

Umbrella groups

The steady trickle of new umbrella groups catering to different sections of the discipline is another of the healthy trends in Australian bioinformatics and bio-IT.

One which popped up recently is the bioinformatics special interest group formed by members of industry association AusBiotech and chaired by the ubiquitous Tim Littlejohn.

An even more recent arrival is the Australian Association of Bioinformatics, catering for workers in the not-for-profit, academic arena, whose formation owes much to the spadework done by Mike Poidinger, head of the Australian National Genomics Information Service (ANGIS).

Another pointer to the sector's growing importance was a November meeting held under the auspices of the National Health and Medical Research Council (NHMRC), the Australia Research Council (ARC) and CSIRO. The heavyweight troika called the meeting to explore ways and means of developing a coordinated national bioinformatics policy for the publicly-funded research environment.

Academia also is perceiving the need for more coordinated programs as the bioinformatics student population grows.

This year will probably generate record numbers of postdoc and graduate students in the discipline and could trigger a determined push for a national graduate level program, according to Prof Mark Ragan, head of computational biology and bioinformatics at Queensland's Institute for Molecular Bioscience.

Private sector optimism

Cautious optimism best sums up the prevailing mood among Australia's small band of private sector bioinformatics companies, whose highest profile players are software and services company Proteome Systems and training company BioLateral.

"We are very happy with the opportunities we see opening up in bioinformatics for Proteome Systems," says executive vice president for bioinformatics Dr Marc Wilkins. Proteome Systems is one of the few Australian companies selling into the international bioinformatics market. The elite group also holds Sydney's Desert Scientific Software, whose founder Dr Neil Taylor has developed a protein structure database and visualisation software package for molecular chemists.

Proteome Systems' ProteomIQ (pronounced proteomic) platform is a high-throughput platform for proteome analysis which integrates the bioinformatic tools needed for storing, manipulating and mining the large amounts of data generated by proteomic instruments.

"We have a platform for proteomics which is underpinned and integrated by our software," says Wilkins. "We have been extremely encouraged by the degree and depth of interest in our products."

The company's bioinformatics focus is on enterprise-level tools because proteomics "is large and complicated and needs that level of computing to handle it."

After laying the groundwork via an early joint venture agreement with IBM, Proteome Systems has pushed forward in the past year by signing a number of pacts with US companies such as Nestles and Charles River Laboratories International which not only stimulate sales of its software platform but promise to generate revenue from contract research conducted using its own toolsets.

"To have a successful international product, you need to be in applied bioinformatics," says Wilkins. "We have things that are very much positioned to help the lab-based researcher acquire more data easily and analyse it."

The company doesn't rule out domestic sales of its bioinformatics offerings but its eyes are firmly fixed on the global scene. "We see enormous opportunities in Japan and the US and also in some of our other neighbours especially Taiwan and Singapore."

Europe is not an arena in which Proteome Systems is well positioned at the moment while China, where the company is having a series of discussions with industry and academic groups, was labelled "an interesting challenge" by Wilkins.

Boutique bio-IT

Private sector companies focused wholly on bioinformatics are thin on the ground in Australia, but growth is slowly taking place. The most significant recent addition to their ranks is boutique bioinformatics consultancy Emphron Informatics, founded by Dr Mervyn Thomas, one of Australia's gifted biostatisticians.

Until he resigned at the end of 2002, Thomas was biotechnology manager in CSIRO's Mathematics and Information Services division and a key figure in the drive to commercialise CSIRO's bioinformatics expertise in the field of proteomics.

His new company is a boutique bioinformatics consultancy, operating principally within the Asia-Pacific region, which offers strategic advice on informatics as well as data mining services for large and complex biological and biotechnology data sets.

Emphron's major client to date is Genetraks, a bioinformatics-based company that called on Thomas to help build a massive database that correlates gene expression data with demographic, clinical, pathological and historical data.

Starting with racehorses, Genetraks believes its technology will produce a sophisticated diagnostic tool for assessing the health of the animals in order to improve their performance.

The impressive commercial potential of the concept became evident in late 2002 when investors pumped $6 million worth of venture capital into Genetraks.

Emphron's arrival in the market suggests the spectrum of commercial opportunities for bioinformatics specialists is growing in Australia and could mark a milestone on the market's march toward maturity.

But the domestic bioinformatics market is a bit like a doughnut with a hole in its middle. The missing bit is the lack of demand from biotech companies for bioinformatics services and software.

"The missing ingredient is more commercial activity involving biotechs," says Tim Littlejohn.

"More biotechs attracting more investments produce more data that requires bioinformatics to turn it into high value intellectual property that can then be commercialised."

He calls "fantastic" the investment by Genetraks in molecular biotechnology and functional genomics and its use of bioinformatics to convert data into valuable IP.

"But what we need is 10 to 20 more companies like Genetraks that consume bio-IT services. I am talking about biotechs working with molecular biotechnology (such as) someone doing koala genome surveys or Great Barrier Reef biodiversity surveys."

Without a supply of biotechs engaged in that type of work, then the population of Australia's private sector bio-IT companies is doomed to remain tiny, Littlejohn says.

One hopeful avenue out of the dilemma leads across the Tasman to New Zealand, he suggests.

Several NZ biotechs including genomics company Genasys and another entity called Vialactia are producing "serious amounts of data -- buckets of commercial high volume biotech data," says Littlejohn.

"The fact is they have been in the trenches creating large, complicated datasets for years. They are impressive, advanced and mature and commercially viable at a global level.

He describes the New Zealand companies as quiet achievers whose internal bioinformatics specialists are "as seasoned as those in the US and Europe" and who have much to teach their Australian counterparts about investing in data production and deploying bioinformatics to create high value IP, Littlejohn says.

"From the Australian perspective, we need to be looking hard across the Tasman to co-ordinate our activity with those kinds of datasets.

"In return, Australia might help New Zealand solve certain database access problems related to substandard Internet connectivity", he suggested.

STATES OF PLAY

The level of bioinformatics activity is growing in virtually every state. In Tasmania, a future bioinformatics powerhouse could be spun out of the $40 million Intelligent Island Initiative in the form of the Centre of Excellence for Bioinformatics that is currently taking shape under the direction of chief executive Dr Michele Allan.

Western Australia's premier bioinformatics entity is the Centre for Bioinformatics and Biological Computing (CBBC) located at Murdoch University and headed by Assoc Prof Matt Bellgard.

A front-runner in NSW's efforts to develop a nationally-relevant bioinformatics research centre is the Sydney University Biological and Informatics Technology Centre (SUBIT) under Prof Tony Larkum. Its brief is to develop high throughput research methods to analyse biological and medical data, with an emphasis on phylogenetics, genomics, proteomics and pharmacoinformatics.

However on the basis of research activity over the past 12 months, the bioinformatics Big Four are probably the Victorian Bioinformatics Consortium (VBC) at Monash University in Melbourne, the Centre for Bioinformation Science (CBIS) at the Australian National University, and the bioinformatics units of Walter and Eliza Hall Institute (WEHI) in Melbourne and the Institute for Molecular Bioscience (IMB) at the University of Queensland.