Open Science: an interview with Rob Lanfear

Robert Lanfear, Senior lecturer and ARC Future Fellow at the Australian National University

 

May 2017

________________________________________________________________

Open Science: what is it for you and why does it matter?

Open science for me is a combination of being open about your plans, your methods, and your results. For me, this involves telling people early the directions in which my group is hoping to make progress, then making sure that all of our code and protocols are open source as we write and develop them, and finally making sure that our published analyses are accompanied by sufficient open data and code so as to make them reproducible.

Open Science and Bioinformatics: is there a link?

Not always, but often. I think bioinformatics (and to a broad extent molecular genetics) was founded on open principles (think GenBank) and these continue to be enthusiastically embraced by most in the field. An excellent example is the current excitement around Nanopore sequencing – many labs around the world are developing new bioinformatic methods to deal with this very promising new type of data, and almost universally those methods are available online (e.g. via GitHub) the minute people have written and tested the code. Preprints and publications (both of which have important roles to play!) are far less important to most people in this area than genuine and rapid methodological progress.

What makes sense to resource as a national effort when it comes to Open Science versus local resourcing/support?

At the national level, I think we need to contribute resources to ensure the use and persistence of key databases and infrastructure. Things like DataDryad, FigShare, GenBank, and the European Nucleotide Archive are really at the heart of open data. Beyond contributing national funds to keep these databases running, I think we should fund national training programs to encourage uptake in using the databases, and that granting agencies such as the ARC and NHMRC should provide funds to support open data and open access publications from funded projects.

How would you recommend a novice biologist approach Open Science and where can they find guidance, resources and tools for getting on board?

I think the best way to approach open science is one step at a time. If you read too many blogs you may be put off by the idea that everything from lab books to grant applications should be immediately made open. I think it’s more important to remember that the more open you are, the better (because that way, science can move forward more quickly and more robustly). I think the first step is to aim for open access publishing accompanied by open data. Perhaps the best places to find support is to look around your institution for someone who is already doing some of these things and buy them a coffee. There are endless resources online, but nothing beats hearing a personal take from someone who’s already tried it.

What are the top three actions/initiatives you would suggest biosciences domains prioritise to enable Open Science, and what type of support do they need?

Reward open science through prizes and other forms of recognition. Consider open science favourably when hiring and promoting people. And, if you’re a granting agency, require that data and publications that result from work that you’ve funded are made open at the time of publication.

EMBL-ABR is a distributed national research infrastructure providing bioinformatics support to life science researchers in Australia. As such what role could EMBL-ABR play when it comes to Open Science for Australian biosciences?

EMBL-ABR could play a central role in running open science workshops, to help demystify the process and get more people on board. These workshops should be particularly targeted at early career researchers, and at PhD, MSc, and Honours students. These are the people who are most likely to get on board, and who can go back to their labs and start to change their PI’s minds too.

Are there bioscience discipline-specific limitations that required tailored solutions when it comes to Open Science rather than enough common denominators for share resources/tools and solutions?

I think there are enough common denominators. Particularly that in almost all cases, more openness is a Good Thing. One slightly more interesting case is that of long-term studies – there is an argument that requiring full openness could remove the incentive for individual scientists to invest in long-term data collection. However, I think the complaints here are overblown – it’s fairly simple to resolve this by the judicial use of sufficiently long data embargoes.

The one thing you would like EMBL-ABR to do in the future when it comes to connections with existing international efforts is…

Provide competitive travel funds for PhD students to undertake short term (say, three month) visits to overseas labs.

What is the best example you can think of when it comes to Open Science in the biosciences and did bioinformatics play any role?

Nanopore sequencing is the best example I can think of. This is an exceptionally fast-moving field, in which the instrument and data themselves change almost every month. To be at the cutting edge of this field requires almost daily updates on everything from lab protocols to the final data analyses. Traditional publication cycles are far too slow to be useful here. Instead, the community is fantastically open. People are sharing their protocols on sites like www.protocols.io, writing detailed blog posts on tips and tricks for analyses, releasing bioinformatics code early and often on GitHub, and posting preprints as soon as an analysis is complete. Bioinformatics plays an absolutely key role here.

____________________________________________________________________________________________________

Biosketch: Rob Lanfear is a senior lecturer and ARC Future Fellow at the Australian National University. His research focusses on molecular evolution and genomics, and he also teaches ethics. He has an enduring interest in open science and likes to encourage it wherever possible. Having started off his academic life in EvoDevo, he says he struggled with the less-than-open nature of that community, in which the race to publish and carefully guarded experimental methods are the norm. He has enjoyed the move across to bioinformatics due to the culture of openness and global collaboration. Rob believes that we can make incredibly rapid progress by (mostly) selflessly sharing ideas and methods as we come up with them, rather than waiting for them to be published.