The genetics and genomics researchers apply a range of molecular and quantitative techniques to investigate economically important traits, traits with an effect on biomedical conditions and inherited disorders. Core areas of research focus on the livestock and aquaculture industries.
Other areas of genetic research include the evolutionary origin and phylogeny of domesticated animals, building comparative genomic maps between diverse animals species.
A broad range of molecular biological techniques, including, DNA sequencing, genetic marker analysis, gene expression analysis, high-throughput phenomics and sophisticated statistical and bioinformatic methodology development, are routinely applied in achieving these objectives. In collaboration with colleagues in reproduction we are exploring the use of gene editing to develop large animal models for disease. Much of the work lies at the interface of molecular and quantitative genetics and the group maintains strong quantitative skills relevant to the discovery and application of molecular findings as well as to basic genetic improvement.
We are working in close collaboration with colleagues in other schools and researchers at national and international institutes (e.g. James Cook University, UC Davies (USA), CSIRO, Department of Primary Industries NSW, Lincoln University (NZ))
A high definition network visualization approach to detect fine-scale population structures from genome-wide patterns of variation
Markus Neuditschko, Mehar S Khatkar and Herman W Raadsma
What is NETVIEW: NETVIEW is a fast and efficient way for identifying and visualizing fine-scale population structures from a genetic relationship matrix among individuals/populations. It identifies and visualizes individual ancestry in an effective way for a range of population structure analyses including the identification of founder individuals, family structures and admixture. A detailed description of the method is provided in:
M Neuditschko, MS Khatkar, HW Raadsma. 2012. A high definition network visualization approach to detect fine-scale population structures from genome-wide patterns of variation. PLoS One submitted
NETVIEW is an analysis pipeline which combines three different software tools to generate a high-definition network visualization of population structures. The tools included in the NETVIEW procedure are:
The working steps of the three different software tools including the various input and output formats/styles are described in the following sections with relevant example files:
About SPIN (SPC)
SPC is the main tool within the NETVIEW procedure. This algorithm is a powerful tool to identify population clusters from any relationship matrix and provides the network structure for a range of further analysis and the final network visualization.
The program SPIN supports different input formats including the options Eucl (Euclidian distance), Cor (Correlation) and Jacc (Jaccard) once an option is selected the input matrix is converted to a distance matrix and an appropriate scatter plot is calculated. If all buttons are un-selected the program expects a distance matrix as input. As our example matrix Example Matrix describes correlations between individuals of five different populations, the Coroption is selected before the matrix is loaded into the GUI using theLoad button.
After loading the relationship matrix in SPIN sorter, press Run SPC button as shown in the following image 1 which will prompt to set SPC parameter (set K=10 and min size=2) as described in the paper. Pressing the OK button performs the clustering and generates the various output files. Image 1
SPC outputs include following images and files:
About network analysis tools (NeAT)
The network analysis toolbox provides a user-friendly web access to a collection of tools for the analysis of networks including format conversion/layout calculation and the calculation of node statistics, including the “degree centrality” statistic. On NeAT page the weighted relationship matrix from SPC W Matrix Example can be transformed into the GLM file (required by CYTOSCAPE) by first selecting the “Format conversion” tab from left hand side panel of NeAT page and then selecting input format option as “Adjacency matrix”, output format as “GLM” and with the option “Edge width proportional to weight” selected. The degree centrality of each individual has been calculated selecting the “Node topology statistics” tab from the left hand side panel of NeAT page. The example of the transformed GML file is given here Example Network and the use of NeAT is shown in following .Image 5.
CYTOSCAPE is an open source software platform for visualizing complex network structures. To visualize the example_network.gml in CYTOSCAPE the file has to be imported: File > Import > Network (multiple file types). Once the network is imported, various visualization manipulations can be performed (e.g. to generate a network as presented in the paper) use the option: Layout > yFiles > Circular or Organic as shown in the following print screen .Image 6.
There are a number of additional options for setting labels, colour and size for nodes and edges. The manipulation of these attributes can be done locally, but we found it more convenient to also import attributes from text files using: File > Import > Node or Edge Attributes. The example files for node colour and node size are given here:
Degree for node size Node fill Color for node colour. To visualize the node size according to the numbers of connections (degree) the degree values have to be associated with node size, which is done by selecting: VizMapperTM > Visual Mapping Browser > Node Size (Double click to create) > Select Degree > Continuous Mapper. Once, all these steps are applied, the final network visualization should appear like Image 7. The file of this network visualization is also provided Network cys and can be opened in CYTOSCAPE.
Please contact Associate Professor Imke Tammen for further information on research projects and funding opportunities.