The overwhelming majority of papers in the Nature Index are authored by researchers at institutions in North America, Europe and East Asia. Researchers from countries in Africa and Central and South America contributed to 5% of the papers in 2014, yet were responsible for less than 2% of the fractional count (FC).
The difference between these figures shows that these regions are highly collaborative. Their researchers most often contribute to papers with international teams and tend to have only one or two representatives from their country in each paper, giving them a low FC. The origins of these collaborations and the relative contribution of the authors varies, yet it is undoubtedly beneficial to the institution or country to have taken part in a high-quality research paper.
The vast majority (70%) of Africa's FC derives from collaborations with non-African countries. Collaborations within the continent are rare, although initiatives are in place to create more.
South African universities are the continent's undisputed leaders. They generate nearly two-thirds of the region's FC and also collaborate widely, both within the continent and with international partners (see 'African network' and 'Africa international'). The University of Cape Town (UCT), viewed by most international university rankings as the continent's strongest research university — and far and away the institution with the largest FC — has the highest domestic collaboration score of all African institutions in the Nature Index. UCT wants to leverage these partnerships and its links to top institutions outside Africa, including Oxford University in the United Kingdom, its most frequent partner in the index, to create international networks.
Such trilateral link-ups take a bit of “midwifery” to set up, but the effort pays dividends, says Danie Visser, deputy vice-chancellor for research and internationalisation at UCT. Disease research yields a more comprehensive picture if it includes results from several countries, and multi-country climate research can say something more meaningful about trends in African weather. For instance, in climate adaptation research, UCT is leading research projects collaborating with African research partners such as the universities of Namibia, Botswana, Ghana and Addis Ababa, as well as with overseas research institutes including the University of East Anglia, University of Oxford, the UK Meteorological Office, the Swedish Meteorological and Hydrological Institute, and California's Lawrence Livermore National Laboratory. The African partners bring expertise and southern hemisphere perspectives on social and environmental issues, and provide opportunities for comparative case studies across the continent, Visser says. The European and US partners bring state-of-the art climate computing resources and expertise, plus a desire to help solve pressing developmental problems.
Visser believes that UCT's collaborative approach is showing results. “Analysts have said that the impact of our science is so high, in both South Africa and Africa, in terms of citations because we have so many collaborations,” says Visser. However, he adds, collaboration between African universities can only increase if the governments put more money into research. “That is one of the most important things that Africa has to take on board.”
Hasnaa Chennaoui with a Martian meteorite.
Collaboration among neighbouring African countries is more common in North Africa, where shared language and culture create strong ties. In the index, Arabic-speaking countries, including Morocco, Algeria, Tunisia, Libya and Egypt, have stronger links to each other than they do with South Africa (see 'African network'). The data also reveal a trend towards more collaborations between France and countries that are her former colonies, particularly with France's leading institution, the National Centre for Scientific Research (CNRS).
Nevertheless, the significance of this historical link is fading, says Hasnaa Chennaoui, an Earth scientist from Morocco's Hassan II University in Casablanca. “There are other new collaborations that are encouraged by scientific programmes, like the [European Union's] Erasmus programme, that open new horizons,” she says.
Chennaoui is the lead author on an index paper1 describing a Martian meteorite that fell in Morocco on 18 July 2011. This was only the fifth such meteorite that people have witnessed falling to Earth, and is thought to have left the red planet around 700,000 years ago. Chennaoui worked with French, Canadian, British and Swiss colleagues on the paper, and says that, in her field at least, collaboration with wealthier countries is more beneficial than partnering locally “with other researchers who have the same analytical problems”.
Developed country researchers have technical capacity that doesn't exist regionally, she says. For the Tissint meteorite, high-tech tests including oxygen isotope analysis and gamma-ray spectrometry of the samples were carried out in the United States and Europe.
“Reliance on a single wealthy partner makes Malawi's college of medicine vulnerable.”
Many African authors in the Nature Index make only a minority contribution to the authorship of a paper. And for some countries, these collaborations revolve around a partnership with one organization. Such is the case for the University of Malawi, which contributed to 11 articles in the index between 2012 and 2014 — 9 of which came from the Malawi–Liverpool–Wellcome Trust Clinical Research Programme, based at its college of medicine and funded primarily by the UK's Wellcome Trust medical research charity. This reliance on a single wealthier partner makes the college of medicine — the only one in Malawi — vulnerable, says Chisomo Msefula, a microbiologist with the partnership. He believes that if the support ended, there would be no immediate way to maintain existing facilities and activities.
Small Multiples small.mu
When considering 2014 African collaborations only, most countries have a weighted collaboration score below 1. The collaboration network within the continent is dominated by South Africa, although there are some regional groups.
Makerere University in Uganda has ten papers in the index in 2012–2014, although its scientists account for a minority of the FC of each article. For example, Lawrence Mugisha from Makerere University is the only African author out of seven on a paper2 that looks at genetic divergence in humans, chimps and their lice. Mugisha says the collaboration started when one of the other authors visited the chimp sanctuary where he worked as a vet. Through this connection, he came to participate as a chimp expert and also collected the lice for the study.
“I work with scientists interested in capacity building in Africa.”
Mugisha says it would be difficult for a Ugandan team to have carried out the work alone: partly for financial reasons, but more importantly because they do not have access to genomic sequencing equipment. He relishes the support he receives from his international partners, and appreciates their contribution. “I have worked with scientists from Europe and America who are very interested in capacity building in Africa,” he says. “Any publication adds respect among the science community and contributes towards my career growth.”
Central and South America
Latin American science has been on the up in the past decade, bolstered by annual regional economic growth of more than 5% between 2003 and 2012. However, in recent years the improvement has eased, with the International Monetary Fund expecting growth to drop below 1% this year.
Steve Gschmeissner/Science Photo Library
South American scientists help decipher the genetics of a pork tapeworm.
The region is still emerging in terms of science output, but countries in Central and South America have a stronger internal research network than those in Africa, particularly among the region's star performers: Brazil, Chile and Argentina, with Mexico close behind (see 'Towards self-sufficiency'). The majority of its collaboration score for 2012-2014 is derived from its international collaborators; the region's own contribution is in the minority. Research also tends to stem more often from extra-continental collaborations (see 'International links').
More than 40% of the region's FC in the Nature Index comes from astronomy papers. This is not surprising given that it hosts important facilities including the European Southern Observatory's Very Large Telescope and Gemini South, the twin of Gemini North in Hawaii. Most of the astronomy articles are written by a mix of scientists from inside and outside the region, many of the latter originating from the science powerhouses of North America and Europe — who helped build the instruments. “This will probably be the case for a long time,” says Felipe Braga-Ribas from the Federal University of Technology-Parana in Brazil, who was the lead author on a paper3 in the 2014 Nature Index describing rings surrounding the asteroid Chariklo. But, Braga-Ribas anticipates self-sufficiency will prevail eventually. The number of scientists in Central and South America is rapidly growing: for example, in Brazil they doubled in number between 2000 and 2010. This should, he says, lead to more intracontinental groups, and more publications without developed-country partners.
Small Multiples small.mu
Central and South American countries in 2014 have a stronger internal network than Africa, with the larger weighted collaboration scores shared across several countries rather than concentrated in one.
To explore these graphics in more detail online, visit: go.nature.com/VQAEi3
The stronger internal connections in this region evolved partly as a response to the political situation between 30 and 40 years ago, says Jose Luis Nilo Castellón from University of La Serena in Chile. Nilo Castellón was the lead author of an index paper4 from a Latin-only group, with members from Chile, Argentina and Brazil, that published an article looking at X-ray luminosity in galaxy clusters using data from the Gemini telescopes.
He explains that the political unrest and lack of funding during the 1970s and 80s made research difficult. “Scientists during those decades made a tremendous effort to think about a better future for Latin American science,” says Nilo Castellón.
Box 1: The Russian connection
Most Central and South American institutions in the index have strong links with countries like the United States, France and Germany. Spain, a former colonial power, is the second largest collaborator for Argentina and Mexico, and the fifth largest collaborator for Chile and Brazil. A notable exception is the University of Sao Paolo (USP) in Brazil, which has the Russian Academy of Sciences (RAS) as its second biggest collaborator, after France's CNRS.
This unexpected connection appears to stem from a handful of influential Russian professors who are based at USP. One of the first to arrive, in 1992, was Dmitri Guitman, a physics professor from Siberia. Guitman has since trained many PhDs and postdocs from Russia and elsewhere. One of these, Maxim Dnovnikov, is back in Russia after three years at USP. While in Brazil, he continued to collaborate with his former Russian colleagues, reinforcing the USP–RAS relationship. “I didn't want to interrupt my previous collaborations, since post doc is a temporary job,” says Dnovnikov. Another factor is the low salaries in Russia, which drive scientists to seek opportunities outsid. However, with Brazil's economy flagging, he suggests such opportunities in the region might become rarer. L.N.
“They created networks of contacts, designed institutional frameworks and lines of development. The Latin American collaboration that we see today is the result of that effort.”
Mexico has solid connections with Brazil, Chile and Argentina, as well as to other countries in Central America. Alejandro Garcia Rubio, a biotechnologist from the National Autonomous University of Mexico (UNAM; Mexico's strongest institution in the index), is one of the chief authors on a paper5 on the genetic sequences of four tapeworm species.
Garcia Rubio also observes that the region is growing in independence: “Collaboration within Latin America is becoming more common than it was a few years ago.” He worked on the tapeworm paper with colleagues from Latin America as well as from countries including the United Kingdom, United States and Switzerland, and says there are benefits to working with both types of partner. On one hand, local researchers easily understand each other's constraints, he says, such as small budgets, long delivery times for reagents and having more junior scientists on board. Researchers in rich countries have access to more resources, are better connected and have a better track record of publishing in high-impact journals, which he says benefits the collaborating scientists from poorer countries. L.N.