Jac Charlesworth PhD: Senior Research Fellow – Computational Genomics, Menzies Institute for Medical Research, The University of Tasmania
How did you come to be working in bioinformatics?
I did a PhD in familial genetics which at that time was through linkage with microsatellite markers. I was studying glaucoma and I had these fantastic really big pedigrees which were way larger than most of the analytical tools could handle. My supervisor had left half-way into my PhD to go back to the US and no one else in Tasmania was doing any kind of human genetics work so I had to self-learn and explore solutions. The upside was that I soon realised that I was not so into the lab work but I loved the data analysis. So I got in touch with the people who had created the software that I thought I could possibly use and I went and did a training course with them at UCLA and loved it! It just so happened that they were collaborating with a group in Melbourne so they came back to Australia every six months so I got to fly across and keep working with them.
So originally you were a geneticist?
Yes, totally. I started as a wet lab PhD student and generated a whole lot of data that I had to analyse to finish the PhD. It was at that point that I realised what I really wanted to do. I have a biochemistry degree (with a little bit of medical science) and a PhD which transformed from wet lab genetics into statistical genetics. My US collaborators offered me a postdoc so I then four years straight in Texas working in computational genetics – no lab work at all (I even had that as a condition of my contract – no lab work!) Then I was lured back to the Menzies Institute to continue those big family studies. There’s some amazing family based complex disease cohorts here.
What are the top three challenges you see for life sciences in the data driven era?
- Access to the relevant compute for the task.
- Access to appropriate information, or more to the point, taking the mass of information that is out there now and filtering out the good stuff – what is appropriate for my data, has this been done before and how can I avoid reinventing the wheel?
- Being able to dedicate enough time to research, collaboration and professional development so that you don’t get stuck in your own little bubble.
Do you think there are different challenges for those who regard themselves as bioinformaticians as opposed to typical life scientists or are the challenges the same?
I think the challenges are slightly different. As a biologist you need to focus back on the biological relevance of the tools you are using rather than getting stuck on ‘this is how the code works’ or ‘this is how the algorithm works’. Some tools are developed to deal with one single problem and may or may not be extendable beyond that problem. I have to be careful that I don’t get too bogged down in the biology and that I’m using the appropriate methods, or that I’m working with the right people to help develop the appropriate methods.
On the flip side of that, the people who are coding have got to make sure that what they are coding is also relevant to what we are trying to get out of the data and what the underlining data actually are ie. we all need context.
Do you think training can help to facilitate awareness or plug the gaps you see in life sciences when it comes to bioinformatics?
Yes, and no. Training is a bit of a double-edged sword. It is fantastic when you already have enough basic information that you know the questions you want to ask and how you want to answer them, but training when you’re already floundering can either potentially take you down a path that isn’t relevant or it can distract you even further from getting to the right information. I have so many biologists telling me they want to ‘learn bioinformatics’ and that this can somehow be achieved in a few days! That right there is a big communication/understanding issue. I think training with a decent level of basic communication and interaction is really important. It also stops you from getting too isolated, too stuck in your particular silo, making sure what you are doing is connected to the bigger picture or the bigger field.
When I say Open Data, what does that mean to you?
It is a very, very small part of what I do.
Is that because you don’t think it is relevant/important or because of the nature of the field in which you work (clinical genomics)?
More-so because in my field the ethical constraints of most of the data that I work on means that open data tends to be just the very end-stage publication.
Are you saying that collaborations and collaborative projects might suffer by an inability to share data?
Yes, but what tends to happen in my field of human genetics is really interesting because most of the collaboration tends to be with the coding, the software design, the platform development and that all tends to be open access / open source. So even though we may not be able to share our individual data sets, we can share our methods and we can share our solutions to problems and this leads to a really interesting research environment. I am not so sure about an environment where everybody just dumps their data and people can go nuts on it.
So what is currently needed in the areas of bioinformatics and life sciences – for Tasmania and then for Australia?
To my mind there’s an issue around how we define what is bioinformatics, which can be very broad or very narrow. Two people can say that they are bioinformaticians and can be doing completely separate things and not even know how to talk to each other, so that is a problem.
The main need I see for bioinformatics and life sciences in Australia is a stronger sense of community. We are often seen as a support science. There’s been a lot of discussion from within the community about how we might go about building a bioinformatics community in Australia and using that to have a voice – both to talk to each other and build collaborative ties but also a voice to funding bodies such as NHMRC to let them know that bioinformatics is a field of scientific research in its own right, not a support science. We need support too.
It is early days yet, but what would you like to see EMBL-ABR become, achieve?
Communication… I would like to see the Resource as a network bringing together the people who have any attachment to bioinformatics or bioinformatics related activities so we can actually talk to each other or know where they are and what they are doing.
Biosketch: Dr Charlesworth runs a computational genomics research group at the Menzies Institute for Medical Research at UTAS. She has a particular interest in complex disease gene discovery using next-gen sequencing in large families, and of using ‘normal’ population variation to inform disease research. She is a statistical geneticist with a background in molecular biology, allowing her to occupy the interface between data analysis and biological interpretation.
Dr Charlesworth is currently hunting for genetic variation underpinning neurological disorders including multiple sclerosis, eye diseases including glaucoma, and other complex traits including cancers, diabetes and brain structural variation. She also manages the computational infrastructure for genomics research at Menzies, and provides bioinformatics training and support.