On scientific collaboration

This is the intro to a collection of essays I edited for the Royal Society, reflecting on the course of a diverse set of scientific collaborations. They appear in a little book(let) recently published. There’ll be a web version at some point, but thought I’d put this up here in the meantime as an appetiser. (yes, I know there haven’t been posts from my archive here for ages – must have got bored with doing that…)

Science isn’t science until you tell someone about it. And you must do it the right way. A paper appears in a journal – part of a system the Royal Society helped establish in the 17th century. That paper is read and refereed before publication, studied and discussed and cited afterwards. Others try and repeat the procedures it describes. Some do new work that extends the results. The whole enterprise depends on researchers working together.

But there’s another, less visible, social side to science that is equally vital. That’s the collaboration that allows the work to get done in the first place. We all know it’s important, in principle. Yes, we celebrate individual genius, but these days that goes with acknowledging that everyone has support networks. The biggest collaborations, like multinational squads probing particle physics at CERN or the human genome project, frequently make headlines. We hear less, though, about the myriad smaller-scale interactions that nourish new science. Nor is it easy to get a sense of how they may change as research develops.

This collection, part of the Royal Society’s work on future cultures of science, is a modest effort to show some aspects of this more clearly. The topic is vast. Collaboration takes many forms, over a range of scales in time and distance. So we’re not trying be comprehensive. This particular slice through the broader landscape of collaboration describes some examples of how it works to allow new science to emerge which would be difficult, if not impossible, to achieve in isolation, and home sometimes it may not work so well. 

Each essay here tells the story of one collaboration. Some are historic, some contemporary, a few still unfolding. They range over fields from research in new vaccines to genetics of diabetes to neuroscience. And even this small sample shows that the dynamics of collaboration are very diverse. If you were advising a young researcher on how to maximise the chances for joining productive collaborations, reading these essays throws up a range of suggestions.

Fran Ashcroft and Andrew Hattersley’s long-standing collaboration on neonatal diabetes brings home the importance of complementary skills – one ingredient that turns people with common interests into a team, who can achieve more together than as individuals. Friendship is important too, and something extra: a spirit of generosity. That extends to believing that it matters more to get the result than apportioning the credit precisely between collaborators (easy to say, harder to do).

Like several of the teams here, they also comment on the importance of stable, and flexible funding. Ray Dolan and Peter Dayan make a similar point. Developing their joint work on brain imaging and mechanisms of reward benefitted from the fact that both had already secured long-term backing from funders who gave them freedom of manoeuvre. When things are moving, there is huge advantage in being able to appoint a new postdoc from existing funds, instead of waiting for a year for new money. As in some of the other stories here, the role of postdocs in cementing collaborations was also crucial. They are open to mastering ideas from several disciplines, and spending time in other labs to hone vital new skills. Such early experience seems to encourage deeper involvements in diverse disciplines later on: collaboration begets more collaboration.

The Dolan-Dayan partnership also underlines that ideas travel best in people’s heads. Their own collaboration was engendered by working in labs next door to each other. Without that fortunate set-up, a collaboration-seeker has to go to where the right people are. Sometimes, that’s another country. Ben Seymour, working on chromic pain in Cambridge, found that it paid to look for collaborators in a place where the culture encourages people to think like him about how to approach the problem. In his case, it was Japan, where he found a stronger orientation toward application of technology.

Making that work relied on a complex constellation of enablers – a history of exchanges between workers with overlapping interests, a five year fellowship in Japan for Seymour, and, ultimately, use of labs in both countries. All that went along with identifying local mentors and installing support to manage the differences in administrative and research cultures of the two countries. This goes  beyond the common view that working abroad broadens horizons, and shows how much extra work and thought is needed to realise the rewards of this kind of collaboration at a distance. Early career internships and exchanges, and a willingness to tolerate development over quite long timescales were also important.

That kind of long-term plotting is a good fit with problems that will have a similarly long life. Sometimes, though, a collaboration is an ad hoc affair, brought together to crack a puzzle that has just arisen from ongoing research. Steve Miller’s account of the observation of an unexpected spectral line in the atmosphere of Jupiter, and its identification as the signature of the exotic ion H3+, is a case in point. As he puts it, this was a case of self-assembly, adding skills to the team as required, and pulling in new information, until they had the answer they sought. This effort, in the 1980s, ultimately involved researchers in ten countries, relying on the then new facility of email, as well as access to 1980s vintage supercomputers. It is a nice case study of observations opened up by powerful new instrument – spectrometers linked to a telescope – calling on contributions from many other specialties to make sense of an unexpected result.

Decoding the spectral lines from Jupiter involved heavy calculation based on one of the most widely accepted, if not always tractable, theories in science – quantum mechanics applied to small molecules. Persuading potential colleagues to take up theoretical prediction when the predictions derive from novel techniques may be harder. This seems to have been an obstacle to forming collaborations to pursue one possible route to a universal flu vaccine, proposed by Derek Gatherer and Darren Flowers a few years ago. Their approach grew out of the new skills of bioinformatics, and it has been hard to get experimentalists to commit resources to test their ideas.

That’s partly because, as other examples in that essay show, the full collaboration needed to make and test a flu vaccine is very extensive. This is an area where a complex infrastructure needs to be built and maintained – including experts in microbiology, immunology and molecular biology, as well as people skilled in organising clinical trials and those who know how to manufacture vaccine in quantity. Add that some vaccines are aimed at viruses that can only be studied in labs with high containment, and the demands of the endeavour become still more complex. Collaborations like this, which are needed to prepare against major global risks, require committed investment, long-term planning and capacity building.

That investment may be in one country, or several. Rebecca Mileham’s investigation of the search for super-heavy elements shows how work in two countries – the US and USSR – mired in political conflict during the Cold War, proceeded separately, but ultimately came together to bear richer fruit thanks to scientists’ efforts to maintain collaboration across borders. It is heartening to find that the optimistic view – that, when it comes to science, researchers speak the same language as their colleagues wherever they work – proved true here in troubled times.

Lay people can learn that language, too – and some learn it well enough to contribute to research. This is true historically, as Sally Shuttleworth and Chris Lintott recall, and indicates that the potential for scientific collaboration is much broader than considering contemporary, professionalised, science might suggest. And as they go on to show, we are now in a new phase of “citizen science”, ushered in by the internet, which is a recovery of a spirit of wider collaboration with widely dispersed, lay observers that was important in a broad swathe of past science. Some of these efforts, involving thousands or even tens of thousands of non-scientists, are probably the largest research collaborations ever mounted.

Involving other people on that scale may enable work that would be hard to achieve any other way. But we must keep room for the essence of small-scale collaboration: people talking incessantly about how to make sense of things. The main thing one of the most scientifically significant collaborations described here needed was a room for two people to share. The two in question, Francis Crick and Sidney Brenner, collaborated intensely as office mates at the MRC Labs in Cambridge for 20 years. Their contribution to science was forged through talk – but a particular kind of talk. As Brenner put it, “it was two people’s minds playing on each other”. That kind of relationship between equals affords an opportunity for uninhibited criticism that helps sift the important ideas from a continual flow of harebrained notions. The lessons of this celebrated partnership may be harder to apply more broadly – both were brilliant, they had an intense and close friendship, and molecular genetics at the time was at a point where enormous problems could be cracked by executing cleverly devised experiments, relatively cheaply, as long as they were guided by inspired theoreticians. It does highlight, though, as Matthew Cobb concludes in his examination of their work, that unlimited time for no holds barred discussion is not a luxury, as it sometimes seems today, but can be the most important spur to progress.

In the end, as the Crick-Brenner history reminds us, collaboration is personal. But it is influenced by the conditions in which research work is done. Understanding those conditions better can underpin the formation of teams with a stronger chance of advancing new work than individuals working alone. We hope this booklet helps inform some of those conversations.



Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s