The Bioinformatics Hub has a keen interest in finding and supporting students with an interest in bioinformatics. Please contact us for information on any future projects.
Ning Liu - Masters of Biotechnology
Developing computational tools for chromosome conformation capture data (4C-seq, HiC, Capture HiC) analysis
Co-Supervisor: Simon Barry, Robinson Research Institute
"The intracellular chromosomal conformation is dynamic. Interactions among genes and regulatory elements have been shown to occur across the genome and play an important regulatory function in gene expression. Using chromosome conformation capture techniques such as 4C-seq, HiC and Capture HiC allows scientists to investigate a specific gene and establish a contact profile within the genome. However, most of the tools for 4C-seq, HiC and Capture HiC data analysis lack a thorough examination and comparison using simulated data, which may identify weakness in these procedure. In our project, 4 4C-seq methods, 18 HiC methods and 2 Capture HiC methods are tested against two dataset of 4C-seq and HiC, one published and one unpublished, as well as against to simulated sequence data."
Jacqueline Rehn - Masters of Biotechnology
The role of DNA damage in ancient metagenomic studies
Co-Supervisor: Laura Weyrich, Australian Centre for Ancient DNA
Studies of the human microbiome have demonstrated that alterations in the diversity and distribution of bacterial populations in the body can result in dysbiosis and disease. While research in this field has helped to understand the aetiology of several complex disorders, questions about what constitutes a healthy microbiome and how modern lifestyles have affected these symbiotic relationships remain. Shotgun sequencing analysis of dental calculus taken from ancient remains can address some of these questions by providing a snapshot of the oral microbiome throughout time. However, authentication of these results is difficult given the complexity of distinguishing between sequences of ancient bacteria trapped in dental plaque and modern environmental contaminants. Ancient DNA from mammals has been shown to have characteristic patterns of damage that can be identified using bioinformatics tools and used to authenticate DNA sequences as being ancient in origin. Bacteria are assumed to demonstrate similar patterns of damage but little research has been conducted to confirm that this is the case. Furthermore, it is unclear how sequencing errors introduced due to the damaged nature of recovered DNA might affect the accuracy of taxonomic profiling. In our project, we aim to use simulated data to investigate the impact of DNA damage on accurate profiling of bacterial populations in metagenomic samples and investigate differences in the type of damage that accumulates in human and microbial genomes.
Melanie Smith - Honours Student
Predictive models for Pregnancy complications using -omic data integration
Co-Supervisor: Claire Roberts, Robinson Research Institute
Pregnancy and childbirth are a time of elevated health risk in a woman’s life. Pregnancy complications, such as preeclampsia, are a major source of maternal mortality and morbidity, and children born from preeclamptic pregnancies are more prone to neurological problems, stroke and diabetes mellitus, and hypertension. At present there is no clinically significant prediction model for the development of preeclampsia. The aim of this project, is to develop non-invasive, computational methods and tools to identify subsets of otherwise healthy, nulliparous women, in early pregnancy, with a clinically relevant risk of developing pregnancy complications such as preeclampsia, Pre-Term Birth, and Small for Gestational Age newborns. These will be developed using the R programming language in conjunction with a MySQL database. Using existing datasets such as SNPs, DNA methylation, RNASeq and micro/macro nutrients from the Screening for Pregnancy Endpoints (SCOPE) project, we will integrate multi-dimensional data to identify biomarkers of interest, and attempt to create a clinically significant prediction model.
Nhi Hin - Master of Philosophy
The molecular signature of Alzheimer's disease
Principal Supervisor: Michael Lardelli, School of Biological Sciences
Alzheimer's disease is a complex neurodegenerative disease and the most prevalent form of dementia in Australia. Unfortunately, the causes and progression of Alzheimer's disease are still not well understood. Most patients with Alzheimer's disease have a form of the disease called 'sporadic Alzheimer's disease', meaning that there is no identifiable cause of the disease. This makes it difficult to study exactly why Alzheimer's disease arises, along with the molecular processes involved in its progression. However, a small subset of Alzheimer's disease patients have 'familial Alzheimer's disease' due to genetic mutations passed down through their families. Many of these mutations have occurred in the PRESENILIN genes PSEN1 and PSEN2, which encode the Presenilins, proteins which are highly involved in the progression of Alzheimer's disease. Although most patients with Alzheimer's disease have the sporadic form, patients with either the sporadic or familial forms of Alzheimer's disease show highly similar brain pathology and clinical symptoms. Consequently, understanding the functions of genes like the _PRESENILIN_ genes in familial Alzheimer's disease is likely to impart valuable knowledge about the causes and progression of sporadic Alzheimer's disease. My research involves comparing RNA-seq transcriptome data from two different zebrafish genetic models of familial Alzheimer's disease developed by the Alzheimer's Disease Genetics Laboratory. Each of these models has a different mutation in the zebrafish psen1 gene. Even though these mutations are different (one results in a truncated Presenilin protein while the other results in a small deletion which does not otherwise disrupt the protein), humans with either of these mutations would similarly develop (familial) Alzheimer's disease. By exploring the overlap between the dysregulated genes and molecular processes in both of these models, it may be possible to establish a molecular "signature" common to Alzheimer's disease cases.
Justin Bogias - Honours Student
Effects of epigenetic modifications in response to the external environment on the patterns of gene expression in Vitis Vinifera.
Co-Supervisor: Carlos Rodriguez-Lopez, School of Agriculture, Food & Wine
It is a widely accepted idea that the composition and quality of grape berries, and consequently of the wine that is produced from the grapes, is largely influenced by the surrounding environment, or terroir, in which that grapevine is planted. Although much literature has been published which focuses on the effect of the terroir on the grapevine, there are still many large gaps in knowledge regarding the subject. External environments can influence organisms through epigenetic modifications to the genome which have effects on the expression of genes as a consequence. DNA methylation of cytosines is one epigenetic process which alters gene expression, by silencing, or in some cases, upregulating expression of genes. The outcomes of DNA methylation are dependent on the context of the methylation. Within the 3D model of the chromatin, methylation events at different loci will elicit various chromosome folding consequences. For example, methylation can influence the formation of 5MB topologically associated domain boundaries which correspond to binding sites for insulator-binding proteins as well as to active transcriptional start sites, eliciting broad consequences for the expression profile of the genome. By taking a computational approach, RNA-seq data and Genome-By-Sequencing data can be integrated to give a gene expression profile for the samples. This data can then be correlated with meta-data containing external environmental information to observe patterns and identify links between gene expression and environmental effects.
Kelly Ren - Masters of Biotechnology
DNA-methylated differences in monozygotic twins discordant for unipolar and bipolar depression
Although monozygotic (MZ) twins share nearly all of their genetic variants, they can be discordant in some particular diseases, such as unipolar and bipolar depression. One possible contributor of this is DNA methylation, which is an epigenetic mechanism influenced by environmental, genetic and stochastic events. Here, our project focuses on the epigenetic analysis of a set of Infinium Human Methylation 450 BeadChip (450k) MZ twins data that has been provided by Flinders University. This data has been obtained from blood samples collected from monozygotic twins aged between 22 to 60. Bioinformatic analysis of this data will hopefully help identify differences in the methylation patterns of depressed and healthy twins. Pathway analysis will then identify the genetic pathways that contribute to depression. Results will be meta-analysed with other MZ twin data from the Brisbane Older Ageing Twins Study with an average age of 70 and the Brisbane Longitudinal Twin Study with average age of 14, as well as similar published studies that are publicly available in order to further discover if age or other factors also play roles in disease progression.
Awais Choudhry - Honours Student
Aristaless related X-linked Intellectual Disability
Co-Supervisor: Cheryl Shoubridge
Intellectual disability (ID) affects between 1 to 3% of the population affecting individuals and their families. Males with ID usually contain a mutation in a gene located on the X chromosome. Aristaless (ARX) mutations is associated with X-linked ID. ARX is frequently mutated causing a variety of phenotypes ranging from severe X-linked lissencephaly with ambiguous gentialia to mild ID with no consistent clinical features. This range of symptoms depends on the location and severity of the mutation. ARX contains four separate poly-alanine (PA) tracts each of which when expanded causes different disease phenotypes. PA1 and PA2 contains approximately 60% of mutations for disease causing mutations in ARX. ARX is a paired homeodomain transcription factor which represses over 800 genes directly. In diseased individuals containing mutations in ARX, over 200 genes are deregulated while many genes expressed regularly, the cause of this remains unknown. We suspect there maybe a correlation between the accessibility of the motif site and its ability to regulate gene expression. Using Hi-C data, Atac-seq and Chip-seq we will investigate the potential causes of certain genes being deregulated by ARX and others remaining functional.