genes

Dissecting the genetic aetiology of complex diseases and traits

Humans differ in a range of traits, and our susceptibility to disease also varies. Studies into the resemblance between family members have suggested that genetic factors contribute to a large proportion of these individual differences. In the last ten years, our researchers have been actively contributing to many of the global efforts to find genes underlying differences in human traits relating to growth, obesity, and behaviour, as well as identifying genetic determinants of blood biomarkers and diseases such as schizophrenia and motor neuron disease.

These discoveries have typically been done as part of large scale international consortia, and our researchers continue to lead and contribute to many of these international research efforts. Indeed, thanks to the joint efforts of the international scientific community, we now know about thousands of genes which are associated with complex traits and diseases. The main purpose of finding these genetic variants is to provide novel insights and valuable knowledge that can be translated into disease prevention, risk prediction, diagnosis, treatment, drug discovery and public health policy. Below we list some examples of our current research projects in the area of genomics of complex diseases and traits.

Current research projects

Trans-ethnic genomic analyses of complex diseases between European and Asian populations

(PI: Dr Beben Benyamin)

Genome-wide association studies have been successful in identifying genes affecting complex diseases, but these studies have been mostly conducted in European ancestry populations. It is not clear whether similar patterns also apply to other populations. With emerging data on populations from non-European ancestry, we can now start answering the questions of transferability across different populations. The aim of the project is to evaluate the transferability of the genome-wide association study findings in Europeans into other populations. This effort will allow us to share the benefits of advance in genomic medicine across different populations. Examples of ongoing trans-ethic genomic projects are on motor neuron diseases and schizophrenia.

Integrative genomic analysis to dissect neurodegenerative diseases

(PI: Dr Beben Benyamin)

As the life expectancy of Australians increases and the population ages, the number of people with aging-related neurodegenerative diseases, such as Alzheimer’s Disease (AD, the most common form of dementia), Frontotemporal Dementia (FTD), Parkinson’s Disease (PD), and Motor Neuron Disease (MND), is expected to increase. Diseases that are characterised by progressive destruction of neurons are estimated to affect close to half a million Australians (dementia - 400,000; PD -70,000; MND - 2,000). No effective treatment is currently available for any of these conditions, and a large number of late phase clinical testing regimes costing billions of dollars, have failed to develop new treatments. It is difficult to diagnose and the clinical presentation, as well as its rate of progression, vary between patients. Novel approaches are now required to answers these challenges. Genetics, with its causal role on phenotypes, has the potential to provide insight and solutions to these pressing problems. The research uses statistical genomic tools and technologies applied on large scale genomics and other omics data such as epigenomics, with a goal to improve our understanding of disease mechanisms, enhance disease management and identify novel therapeutic opportunities.

Estimating genotype-environment interaction using genomic information

(PI: Dr Hong Lee)

Previously we have developed novel statistical methods to dissect the genetic architecture of complex traits that are captured by single nucleotide polymorphisms (SNPs) in genome-wide association studies (GWAS). The aim of this project is to extend this methodology in order to estimate changes in genetic effects at the genomic level with changes in environmental factors, i.e. genomic genotype-environment interaction (G x E). G x E is a common biological phenomenon in complex traits where genotypes respond to different environmental conditions in different ways.

However, it has been difficult to study G x E in human genetics due to the need for a design where the same genotypes are measured across environments or they are expressed longitudinally over a trajectory of a continuous environmental variable. In farm animals G x E has been estimated based on sires’ offspring across environments but this is more difficult in human data where family sizes are smaller and family effects are often confounded with environmental factors. We aim to develop a new method to estimate G x E based on genotype effects at the genome level. This approach is more versatile in relation to the design of the data as most variants are expressed across environments. The outcome of this application will enable a better understanding of genetic responses to environmental changes and how different genomic regions contribute to phenotypic variation under changing environmental conditions. Therefore, the project contributes towards the fundamental understanding of biological processes and phenotypic variation.

Novel methods for complex trait analyses based on genome-wide approaches

(PI: Dr Hong Lee)

Methods using linear mixed models and based on genome-wide genotype information have been developed to dissect the genetic architecture of complex traits and predict genetic merit or risk in various species including plants, livestock, mice and humans. Although successful for prediction within breeds or populations, the methods are still relatively simple from a genetic point of view and often lack power for predictions across populations. We propose to extend the existing methods by exploring more systematically differing genotypic effects in different genetic background and environments.

We refer to this as modelling a ‘dynamic genetic architecture’ and will determine the utility of applying such models by evaluating improvement in genomic prediction accuracy. Genomic predictions are generally based on heterogeneous training populations with individuals having a varying degree of genetic similarity with the target individuals to be predicted. Closer relatedness leads to sharing of larger genome segments and therefore higher prediction accuracy, even for additive genetic models. Varying genetic effects in different genetic backgrounds (due to epistasis) and environments (due to genotype by environment interaction) will further affect predictive power in relation to genetic relatedness.

The aim of this project is to further develop linear mixed models for genome-wide prediction by utilising the information from the dynamic genetic architecture of complex traits, i.e. the changes of genetic characteristics and effects with varying environment, genetic background and genetic distance, and to develop efficient computational tools for such analysis. The outcome of this project will enable a better understanding of the ability to predict genetic effects, depending on population and environment, and to understand the value of heterogeneous information sources in genomic prediction. This will affect design and accuracy of genomic prediction methods.

Large scale collaborations discovering genetic influences on growth and human behavior

(PI: Professor Elina Hyppönen)

The foundations of our health and wellbeing in adulthood are in large part laid during the early years of development. For example, we know that childhood obesity tends to track to adulthood, while low birth weight has been associated with increases in the long term risks of many metabolic diseases. Therefore, if we can better understand the genetic and environmental influences which affect aspects of health and wellbeing in childhood, this can also help us to develop better preventative strategies to improving health later in life.

Since 2009 Professor Elina Hyppönen has been involved with international consortia efforts to disentangle genetic determinants which specifically affect traits in early childhood and adolescence. Notably, Early Growth Genetics Consortium (EGG) is focused on identifying genetic determinants affecting growth-related phenotypes and covers traits such as birth weight, childhood obesity and pubertal development.

Another related effort is the Early Genetics and Lifecourse Epidemiology (EAGLE) consortium, where we aim to identify genetic factors which affect childhood behaviours, including aspects of psychopathology and neurodevelopment. These long-term collaborative efforts continue, with exciting future prospects to discovering novel insights into disease mechanisms and optimally, strategies of prevention.