Dr Gareth Price, Head of Computational Biology, Queensland Facility for Advanced Bioinformatics (QFAB)
25 May 2018
What role does bioinformatics play in your work, and why is bioinformatics important?
Bioinformatics is core to my work but more importantly core to the work of the Australian researchers that my roles in QFAB and Galaxy Australia support. As Head of Computational Biology at QFAB I help researchers from all over the country to expedite their research by providing analyses on their datasets. It is my experience that even with falling costs of data generation, particularly in genomics, one of the biggest shortcomings in the data journey is the ability to maximise value from cost by performing the most appropriate (best practice or bespoke) interrogation of the data. This is where the value of good bioinformatics becomes critically important to the researcher. Not only is it the choice of analytical tool for the question but also knowing the best resources within Australia on which to perform the analysis. Without the combination of best tool and most appropriate computer hardware the true potential in a dataset may never be realised. So to answer “why is bioinformatics important” one must first acknowledge that researchers are the best placed to answer their own research question – they know the literature and they know the best experimental model to test their hypotheses. However if they do not know the best informatics approach to extract the evidence from their data then the maximal potential in the data will never be realised. It has been my genuine honour to help researchers navigate the analysis options for their data and provide back to them results to advance their research goals.
Is there a bioinformatics skills shortage in Australia and how do you experience this in your own research or organisation?
Bioinformatics is still a relatively need field of study, as such the skills shortage can be observed at multiple levels: undergraduate course content, HDR students undertaking new studies and researchers applying new techniques to their research questions. There is a need to simultaneously tackle the upskilling at all levels. Interestingly, the rapid capacity to generate data in the life sciences has led to a series of skills shortages beyond bioinformatics. Data storage, data management, data traceability and metadata use need to be taught at all levels of study and research activity as necessary practice to maximise the value for each dataset. Thankfully we have a guiding hand in the FAIR principles, for the Findable, Accessible, Interoperable and Re-usable themes that can be applied by researcher to all datasets.
This year I have taken on the role of Program Manager and Service Manager for the Genomics Virtual Lab and Galaxy Australia. Galaxy Australia is ideally suited to help address the skills shortage in bioinformatics analysis. It offers an approachable GUI interface for analysis tools, reference datasets and compute hardware. Galaxy Australia allows, through free registration, access to a maintained toolset of common bioinformatics tools for the manipulation, translation and filtering of data plus the visualisation of data on a genomic level. Without the need for skills in command line tool execution users can run single tools but where the power of Galaxy Australia comes to the fore is in the ability to chain and brain tools in a workflow that can be run and rerun by the user. Plus workflows and data histories can be shared with others in Galaxy facilitating collaboration.
Why has training and skills development in bioinformatics become so important in life science research?
In many ways having skills in bioinformatics for the researcher should be viewed as just having the skills for any modular component of experiment flow. We wouldn’t or we shouldn’t let researchers unskilled in a laboratory technique perform that technique, at least not without expecting poor results or maybe endanger themselves and we shouldn’t waste time and effort using the wrong tools on good data. Since the types of data being generated to test research hypotheses in the life sciences have grown in volume and complexity, it is only logical that the skills required to do the analysis need to grow in parallel. EMBL-ABR’s efforts to support the upskilling of researchers is essential if we are to make bioinformatics a routine part of the flow from experimental design to conclusion.
How do you see such a skills shortage impacting health outcomes, in particular, in Australia?
This is a very interesting question, to draw a link between skills shortage in the life sciences, the use of bioinformatics and the impact on health outcomes in Australia. I say interesting because I firmly believe the link between the two is genuine but complex. At one level the integration of genomic analysis in clinical management has flowed, often at a startling pace, smoothly from genetic testing in clinical practice. However, the added resolution of genomic testing has sparked questions that were of minimal scope at a single gene level but significant at a genomic level, such as incidental findings (the confirmation of a variant outside of the clinical request but carrying significant pathological outcomes). Conversely certain genomic tests have also reduced risk, an excellent example being non-invasive prenatal testing where the risks of the foetus carrying particular genetic conditions can be ascertained for the mothers peripheral blood, not from the womb itself, where sampling contains an inherent risk of inducing miscarriage.
In the prescription of pharmaceutical agents the inherent genetic capacity to utilise the agent is rarely factored into the dosage. In part this is due to the skills shortage in interpretation of the genetic measurements (by NGS or arrays). In part is also due to the greater community level lack of knowledge in the benefits that genetic testing can provide over a person’s lifetime, not just the risks to health insurance but rather the benefits to health care management.
Finally in oncology, whole cancer profiling at the genetic or transcript level plus the newer technologies allowing for the assaying of single cells, has further highlighted the microcosm that is a tumour. To make the best use of single cell sequencing, new analysis methodologies had to be developed and their implementation is still in the hands of the skilled few. This serves as an excellent example of a skills shortage that can have a great impact on health outcomes, by the explicit and fine-grained description of a tumour with the view to describe targeted treatment options.
In a very timely manner the activity of the various Australian Genomic Health Alliances and the recent Federal Budget announcements for greater investment in eResearch and genomic research show the recognised need to maximise the benefit from genomics in health care.
What difference might be made to our performance in markets such as health, agriculture, education and biotechnology, if we find ways to address this skills shortage?
The temptation is to try and oversell the importance of bioinformatics to all these markets, the reality is no oversell is required – bioinformatics will, in my opinion, help drive all the major advances in health, agriculture, education and biotechnology we’ll see in the next few years. How specifically it will achieve this involves some crystal ball gazing to nominate the most likely examples. Instead, drawing on where genomics is currently changing markets and projecting onto the two major cohorts of researchers and lay persons, we can surmise that the increased knowledge of bioinformatics will drive innovation from labs to the community and back from the public to researchers, through policy change, medical requests, direct to consumer testing and demands for cheaper and safer products.
The ongoing description of both important agricultural species as well as endangered species at the reference genome level requires extensive bioinformatics, with outcomes for better breeding programs and or species management.
Education and bioinformatics should happen at all levels of schooling and beyond to the general community. Thanks to TV shows like Who Do You Think You Are? (SBS) and TV advertising from Ancestry.com, the public vocabulary around genomics has become more common place. By educating students in the life sciences and the role that bioinformatics can play in answering complex biological questions, we help inform students about the STEM potential for their ongoing education. I count myself fortunate enough to recently have been involved in such an education event, where at Redbanks Plain High School selected students with an interest in STEM were involved in a full day event and I was able to present just how far knowledge of computing and knowledge of biology or the combined knowledge of bioinformatics can take us.
What role do you see for the EMBL-ABR network?
The EMBL-ABR network in my view is ideally placed in two ways to play a role in the continuing upskilling of Australian researchers. Firstly, through the development and deployment of training material within Australia, EMBL-ABR can disperse knowledge around Australia in an efficient manner. This is why Galaxy Australia is making use of the EMBL-ABR Train-the-Trainer and hybrid training models to educate researchers on all the features of Galaxy Australia. Secondly, as EMBL-ABR is a collection of nodes spanning Australia it is very well placed to hear of gaps in people’s skill sets. As new technology platforms become available and offer new research questions, the EMBL-ABR network can be the “ear to the ground” to guide the development of new training materials.
Biosketch: Gareth Price is Head of Computational Biology at the Queensland Facility for Advanced Bioinformatics where he helps Australian researchers with bioinformatics analysis ranging from best practice NGS data processing to highly bespoke genomic questions. Gareth has for 2018 taken on the role of Program Manager and Service Manager for the Genomics Virtual Lab and Galaxy Australia platforms. Prior to this Gareth was a Genomics Scientist for over 15 years. He has been involved in experimental design, assay performance, data QC, data analysis and data interpretation from early printed microarrays, to cartridge based GeneChips through to multiple Next Gen platforms. These works have involved a variety of model organisms from microorganisms, fruit flies, mice to humans.
Gareth’s view is that research, clinical research, and healthcare are at their best when coupled with the most accurate, highest throughput and innovative technology and analyses. He uses this view to motivate the use of innovation to reduce the time between data generation and data summarisation, ready for the important phases of data interpretation and result discovery.