SCIENCE AT THE SHINE DOME canberra 6 - 8 may 2009

Early-career researchers

Thursday, 7 May 2009

Moran award

Dr Melanie Bahlo
The Walter and Eliza Hall Institute for Medical Research

Melanie Bahlo graduated with first class honours in statistics from Monash University in 1992. Her PhD studies focused on gene conversion models in the coalescent under the supervision of Professor Bob Griffiths. For her first postdoctoral position (1997 to 1999) she worked on migration models in the coalescent. In 1999 she moved to the Walter and Eliza Hall Institute for Medical Research (WEHI) to work with Professor Terry Speed in the Genetics and Bioinformatics Group. This move saw a shift in her research interests from population genetics to statistical genetics and bioinformatics. She currently holds an NHMRC Career Development Award and is head of her own research group within the Bioinformatics Division at WEHI. Her interests cover statistical applications in genetics including statistical genetics, population genetics and bioinformatics. She has been involved in the discovery of several new genes and has developed statistical methodology for genetic mapping.

Hunting for genes involved in disease

Genetics is a rapidly evolving field where profound leaps in technology periodically revolutionise our ability to look at genetic data. Taking advantage of these advancements requires understanding of genetics, knowledge of the technology and above all, statistical thinking. This combination can lead to rapid progress in the understanding of the genetic mechanisms that cause human diseases. Humans have around 20,000 genes in a genome of length 3 billion base pairs. It is a numerically daunting challenge to identify the faulty gene in a family with a genetic disease of unknown cause. Action myoclonus–renal failure (AMRF) syndrome is a rare, deadly, recessive genetic disorder. The gene causing this disease could not be identified using standard family-based approaches since no families with more than one affected person could be found. We obtained DNA from three unrelated individuals with AMRF, their parents and unaffected siblings. We generated genetic data from 50,000 genetic markers with the then new high throughput technology, a SNP chip. By generating an empirical distribution from a publicly available dataset and comparing our affected and unaffected individuals against it, and then overlapping the results from both, a genetic region containing about 50 genes was identified. Next, the gene expression levels in blood samples from 2 families were measured using expression microarrays. By only looking at genes in the candidate region we were able to identify a gene that had reduced expression in the affected individuals. Sequencing identified four different mutations in the gene.