|
|
|
|
|
|
|
|
|
| CSIRO | SOLVE | Issue 9 | NOV 06 | |
|
| ||
|
ARTICLE
MEDICAL RESEARCH:
Genes Help Tumour Diagnosis By Jason Major
Advanced statistical tools are being used to identify genetic markers that will allow doctors to more accurately diagnose different types of brain tumourStatisticians and medical researchers are teaming up to pinpoint genes that can act as diagnostic markers to better distinguish different brain tumour types as the first step in increasing the long- term survival rate of patients. The Kolling Institute of Medical Research at the University of Sydney is working with CSIRO statisticians who have developed advanced statistics to analyse the activity of tens of thousands of genes placed onto a tiny, wafer-thin chip called a microarray.
Kolling Institute postdoctoral scientist Dr Kerrie McDonald says the use of advanced statistics has helped advance the institute's brain tumour research by up to five years. "By isolating only the genes of interest, we have saved years of additional laboratory analysis," she says. There are a number of brain tumour types, collectively called gliomas. They range from low grade to aggressive and highly malignant. At the moment, pathologists determine the tumour type with a microscope. They make a diagnosis based on how a cell looks and behaves. "Unfortunately, this technique cannot help us identify useful molecular targets to better understand the mechanisms of brain tumours or guide us when selecting the most appropriate treatment protocol," Dr McDonald says. Since the institute and CSIRO started this collaborative project, researchers have found three key genes associated with the more aggressive types of brain tumour. These can now be used as accurate diagnostic markers. This in turn should lead to more appropriate and potentially life-saving treatment. The three genes are subject to a patent based on their use in a diagnostic setting.
The advance that has been achieved is that while traditional statistical analyses used for microarrays can identify significant genes, the resulting list can number 100 or more. All of these genes then have to be biologically validated by the medical researcher to ensure that any genetic effect is real and not an artefact of the microarray platform or processing. "The problem is choosing which genes to validate as there is neither the time nor money to validate this many," says CSIRO statistician Maree O'Sullivan. However, CSIRO's microarray analyses identify a small enough number of potentially significant genes to allow all to be validated in the laboratory. "The advantage is the ability to identify small gene subsets," Ms O'Sullivan says. "It is the genes within these subsets that discriminate the different brain tumours. "And because our analyses can be applied to any microarray platform, it has also proven valuable in other research areas such as agriculture. For example, we are now investigating genetic resistance to amoebic gill disease in Atlantic salmon that costs salmon farmers millions of dollars each year," she says. In 2007, CSIRO and the Kolling Institute will try to find markers that can predict patient survival, and genetic signatures that will predict how a patient might respond to a particular treatment. For further information contact: |
 IN THIS ISSUE
|
| Home | About Us | eSubscribe | Links | |
|
|
||
|
|
|
Back to main