For a very biodiverse nation, Peru has alarmingly patchy knowledge of its plants. Whereas regions such as Machu Picchu are well documented, vast corridors between the Andes and the Amazon Basin remain scientific blind spots.
Competing Interests
The author declares no competing interests. More
“I’m a botanist and wildlife expert working as a research coordinator in the Gashaka-Gumti National Park in eastern Nigeria. Unfortunately, the park has suffered from decades of illegal logging, poaching, uncontrolled grazing and bush clearing. But since 2017, the charity I work for, called Africa Nature Investors Foundation and based in Lagos, has been restoring the park as a haven for wildlife and indigenous plants, in partnership with Nigeria’s National Park Service.This photo was taken in March, at the end of Nigeria’s dry season. I was riding my motorcycle down a track in the heart of the forest, 15 minutes from our base camp. I saw a striped kingfisher (Halcyon chelicuti) in the afternoon light, and wanted to take a picture of the bird.People and dogs team up to protect sea turtles in Cabo Verde
Login or create a free account to read this content
Gain free access to this article, as well as selected content from this journal and more on nature.com
Access through your institution
or
Sign in or create an account
Continue with Google
Continue with ORCiD More
Each week since 1977, researchers at the Portal Project have monitored how rodents, ants and plants interact with each other and respond to their climate on plots of land in Arizona. At first, the team shared those data informally. Then, beginning in the 2000s, the researchers would publish a data paper, wait several years and then publish a new one with combined old and new data to keep the information current.“Data collection is not a one-time effort,” says Ethan White, an environmental data scientist at the University of Florida in Gainesville, who began collaborating with the project in 2002. New tools have allowed the team to automate and modernize its strategy. In 2019, White and his colleagues developed a data workflow based on the code-sharing site GitHub, the data repository Zenodo and the software automation tool Travis CI, to keep their data current while preserving earlier versions (G. M. Yenni et al. PLoS Biol. 17, e3000125; 2019); so far, the Zenodo repository holds around 620 versions. “We wanted an approach that would let us update things more consistently, but in a way that if someone ever wanted to replicate a past analysis, they could go back and find the precise original data that we used.”Long-term ecological research is not the only area that needs to maintain and update data for future use. Many researchers add to, revise or overhaul their data sets over the course of their projects or careers, all while continuing to publish articles.But despite the need to update and preserve versions of data, there is little guidance for how to do so, says Crystal Lewis, a freelance data-management consultant in St. Louis, Missouri. “There are no standards for repositories; the journals are not telling you how to correct a data set or how to cite new data, so people are just winging it.”Good data-science practice can make the process more methodical. Here are five tips to help alter and cite data sets.Choose a repositoryAlthough it’s easy to place data on personal websites or in the cloud, using a repository is the simplest way for researchers to store, share and maintain multiple versions of their data, says Kristin Briney, a librarian at the California Institute of Technology in Pasadena, who helps researchers to manage their data. “It’ll get it out of the supplemental information; it’ll stop being shared upon request; it’ll stop being shared on personal websites,” on which it can be lost.By the end of this year, US federal funding agencies will require researchers to put data in a repository, with some agencies, including the National Institutes of Health, already implementing the policy. Some journals also require authors to use data repositories. PLoS ONE, for example, recommends several general and subject-specific repositories for its authors, including the Dryad Digital Repository and Open Science Framework.Challenge to scientists: does your ten-year-old code still run?A repository, or data archive, is more than just cloud storage. Repositories provide long-term storage with multiple backups. Zenodo, for example, says that data will be maintained as long as Europe’s particle-physics laboratory CERN, which runs the site, continues to exist. Generally, repositories also promise that archived data will remain unaltered and assign a persistent identifier to data sets so that others can find them.Briney suggests that researchers check whether their funding agency has specific recommendations. There might also be a particular repository for the type of data, such as GenBank for genetic sequences; or a discipline-specific repository for the field of study. Some universities offer institutional options, which usually have the added benefit of technical support. When there is no specific repository available, the non-profit organization the Gates Foundation in Seattle, Washington, recommends generalist repositories, such as Zenodo, Dataverse, Figshare and Dryad.Create multiple versionsFor transparency and accessibility, making a new version when data are added is essential. The alternative — overwriting the old data with the new — makes it impossible to repeat previous analyses or to see how the data have changed over time. Although best practice around creating versions and data alterations tends to focus on future users and scientific reproducibility, the real beneficiary is the researcher, says Lewis. “Three months from now, you will forget what you did — you will forget which version you’re working on, what changes you made to a data set. You are your biggest collaborator.”This is when data repositories come into their own, because many create new versions by default when data are added. Some repositories, such as Zenodo, also mint a digital object identifier (DOI) for each version automatically. “Since the very beginning, Zenodo has provided versionable data with individual DOIs that will take you to a specific version of the data, and also an overarching DOI that will link together all of those versions,” says White. That creates an umbrella link, as well as a mechanism to cite specific versions of the data.Managing versions without a repository is also possible. Researchers who store their data on GitHub, for instance, can use automation to create new ‘releases’ whenever they update their data. They can also create a version of the data set manually, using distinct file names, to differentiate these files from the earlier set, Briney says.Define file names and terminologyBriney regularly helps researchers to wrangle their data. Her favourite tips for data management are to establish a file naming convention, which includes the date (often given as YYYYMMDD or YYYY-MM-DD), and to store files in their correct folders. This is true whether you’re storing data locally or in remote repositories. “It takes 10 minutes to come up with a file-naming convention, everything gets organized, and that way you can tell related files apart,” she says. “It’s like putting your clothes away at the end of the day.”Briney also recommends documenting metadata, explaining the different variables used, and the location of data in the various files and folders. These practices “help you, but are also good for data sharing, because somebody else can pick up your spreadsheet” and understand it.Eleven tips for working with large data setsSabina Leonelli, who studies big-data methods at the Technical University of Munich in Germany, says that researchers should also explicitly document the terminology and queries used to generate and analyse their data. She gives an example of research using a biomedical database: “When you access certain databases, you frame your query” based on current definitions, she says. As knowledge develops, definitions shift and change, and if the specific definitions you used aren’t captured, she says, you might forget the query that originally shaped your data.Write a change log
Login or create a free account to read this content
Gain free access to this article, as well as selected content from this journal and more on nature.com
Access through your institution
or
Sign in or create an account
Continue with Google
Continue with ORCiD More
Download Nature hits the books 11 July 2025Vianet Djenguet is an award-winning wildlife film-maker and camera operator whose work has featured in a number of major nature documentaries. In this podcast, Vianet joins us to talk about his career, how wildlife film-making have changed, and his experiences working with local researchers to capture footage of endangered animals on the new television series The Wild Ones.Music supplied by SPD/Triple Scoop Music/Getty ImagesNever miss an episode. Subscribe to the Nature Podcast on Apple Podcasts, Spotify, YouTube Music or your favourite podcast app. An RSS feed for the Nature Podcast is available too. More
Over the past few decades, evidence synthesis has greatly increased the effectiveness of medicine and other fields. The process of systematically combining findings from multiple studies into comprehensive reviews helps researchers and policymakers to draw insights from the global literature1. AI promises to speed up parts of the process, including searching and filtering. It could also help researchers to detect problematic papers2. But in our view, other potential uses of AI mean that many of the approaches being developed won’t be sufficient to ensure that evidence syntheses remain reliable and responsive. In fact, we are concerned that the deployment of AI to generate fake papers presents an existential crisis for the field.What’s needed is a radically different approach — one that can respond to the updating and retracting of papers over time.We propose a network of continually updated evidence databases, hosted by diverse institutions as ‘living’ collections. AI could be used to help build the databases. And each database would hold findings relevant to a broad theme or subject, providing a resource for an unlimited number of ultra-rapid and robust individual reviews.Adding fuel to the fireCurrently, the gold standard for evidence synthesis is the systematic review. These are comprehensive, rigorous, transparent and objective, and aim to include as much relevant high-quality evidence as possible. They also use the best methods available for reducing bias. In part, this is achieved by getting multiple reviewers to screen the studies; declaring whatever criteria, databases, search terms and so on are used; and detailing any conflicts of interest or potential cognitive biases.Scientists are building giant ‘evidence banks’ to create policies that actually workYet these reviews require considerable resources. Some studies suggest that Cochrane reviews — systematic reviews of specific topics in health care and health policy that meet internationally recognized criteria for the highest standards in evidence-based health care — generally cost more than US$140,000 and take more than two years to complete3,4.It is becoming ever harder for review authors to keep up with the rapidly expanding number of papers. The scientific literature is estimated5 to have doubled every 14 years since 1952.Because each reviewer tends to have access to different publications, and because databases are continually updated, systematic reviews are plagued by reproducibility issues. A study published last year concludes that only 1% of reviews report a search strategy that is fully reproducible6. Furthermore, many systematic reviews unwittingly cite publications that have been retracted, including those removed from the literature because of methodological or ethical issues and fraud7.We agree that AI could be part of the solution to these problems. It could help investigators to conduct reviews more comprehensively and more efficiently — by filtering many more papers, say, or by assessing the entire content of papers instead of just the title and abstract, as human reviewers tend to do as a first step. But one aspect seems to be underappreciated: the degree to which AI — particularly large language models (LLMs) — could exacerbate some of the problems.At this point, little is known about how many scientific papers generated entirely by AI are being published. As announced in March, a scientific paper8 generated by AI Scientist (an AI tool developed by the company Sakana AI in Tokyo and its collaborators) passed peer review for inclusion in a workshop at a key AI meeting. The reviewers did not detect that an AI model had formulated the hypotheses, designed and run experiments, analysed the results, generated the figures and produced the manuscript.Policymakers in China are using evidence synthesis to guide management of the invasive grass Spartina alterniflora.Credit: SIPA Asia/ZUMA Press/AlamyAnd a preprint posted on arXiv estimates that at least 10% of all PubMed abstracts published in 2024 were written with the help of LLMs, on the basis that an abrupt increase in the frequency of certain words coincided with widespread access to LLMs9. That proportion has almost certainly gone up since.Decision makers need constantly updated evidence synthesisEven if LLMs are used widely, it is difficult to separate cases in which they have been deployed to fabricate papers from those in which authors are simply using them to improve their writing10. Yet generative AI is likely to make the production of fake manuscripts easier, irrespective of whether those who use LLMs maliciously do so to further their careers, to manipulate the conclusions of evidence syntheses because of a specific commercial or policy objective or simply to be disruptive. The use of multiple LLMs will also make it more difficult for humans to detect textual fingerprints associated with one particular model.In other words, the use of generative AI is likely to supercharge the already growing problem of paper mills — businesses that sell fake work and authorships to researchers seeking journal publications to boost their careers. It could even replace the paper-mill market, given that fake papers can now be generated in minutes for free.What to do?The Campbell Collaboration (a group of researchers and policymakers dedicated to generating evidence syntheses for economic and social policy decisions) and Cochrane already provide guidance on how to identify studies that have raised concerns or that have been retracted11. This includes checking studies against the Retraction Watch database, which lists retractions gathered from publisher websites, and using the CENTRAL database, a repository for clinical-trial reports that flags retracted studies11. Cochrane guidance also states that the authors of published reviews containing retracted studies should recalculate all results and, while doing so, flag the review with an editorial note or withdraw it and then publish the updated version11.Even now, this kind of reanalysis often fails to happen, presumably because the original review authors have limited resources and little incentive. In one assessment of systematic reviews of pharmaceutical compounds tested in clinical trials, retracted papers continued to be cited in 89% of the reviews one year after the review authors had been notified of the retraction7. With the ever-increasing production of both legitimate and spurious scientific literature, researchers’ ability to maintain an accurate picture of what the data show is likely to be outstripped (see ‘More papers, more retractions’).Source: Data from Open Alex (https://openalex.org/).So, what system might enable the continual and rapid removal — at scale — of fraudulent or otherwise problematic papers from databases?Although not developed with this goal in mind, our work on the Conservation Evidence project — an information resource hosted by the University of Cambridge, UK, to support decisions about how to maintain and restore global biodiversity — has convinced us that a network of AI-enabled, continually updated evidence databases is one possible solution.As part of this project, all of the authors of this article have been involved in developing subject-wide evidence synthesis. The aim here is to identify literature containing information that is relevant to a broad theme. For the Conservation Evidence project, this is the effectiveness of management actions for biodiversity conservation.Evidence synthesis needs greater incentives
Login or create a free account to read this content
Gain free access to this article, as well as selected content from this journal and more on nature.com
Access through your institution
or
Sign in or create an account
Continue with Google
Continue with ORCiD More
The US Fish and Wildlife Service (FWS) and National Marine Fisheries Service (NMFS) are proposing to “rescind the regulatory definition of ‘harm’” in their regulations pertaining to the Endangered Species Act (ESA) of 1973 (see go.nature.com/4jdfgud). The aim is to remove habitat protections and focus only on preventing intentional acts that cause injury to individual animals. This is misguided.
Competing Interests
The author declares no competing interests. More
Nature’s Memory: Behind the Scenes at the World’s Natural History Museums Jack Ashby Allen Lane (2025)Natural history museums are crucial for conservation — and for communicating its importance to the public. But step through the ‘staff only’ doors, and museums look very different. In vaults and laboratories, curators store, catalogue and preserve millions of specimens collected from the natural world — sometimes in controversial ways.In Nature’s Memory, zoologist Jack Ashby explains how all of this works, and the human choices that it entails. Ashby, who studies marsupials and monotremes, is assistant director of the University Museum of Zoology in Cambridge, UK. He is also president of the Society for the History of Natural History in London.In his office, plastered with posters of some of the world’s best nature dioramas, Ashby told Nature about his quest to communicate the importance of natural history museums.Do you have a favourite?One is the Biological Museum in Stockholm. The building looks like a wooden Norwegian church, and it’s effectively one giant diorama that goes up three storeys. It covers all of the Nordic biomes. And it shows that we do have some exciting wildlife in Europe.Egypt is building a $1-billion mega-museum. Will it bring Egyptology home?Another is the National Natural History Museum in Paris. You walk into the comparative-anatomy section and it’s one giant wall with thousands of skeletons all facing towards you, so tightly packed that you cannot walk between them.These probably shouldn’t be my favourites, because they are so old fashioned, but they are stunning.Do museums teach science in a neutral way? In one book chapter, you highlight male biases in specimen collection and display.Natural history museums are amazing, but of course they are built by people, and people have interests and biases. One study, by biologist Natalie Cooper at the Natural History Museum in London and her collaborators, looked at more than 2 million specimens at 5 museums, and found that only 40% of the birds were female (N. Cooper et al. Proc. R. Soc. B 286, 20192025; 2019). For mammals, the figure was 48%, but in some of the mammalian groups, particularly artiodactyls — such as deer and antelope — only 40% were female. In another study, curator Rebecca Machin found that almost three-quarters of the natural-history specimens on display at the Manchester Museum, UK, were male (R. Machin Museum Soc. 6, 54–67; 2008).The numbers are huge, but it’s also about how they are displayed, presented and interpreted. For example, descriptions of male specimens are much more likely to give general facts: this is where the animal lives, how it’s adapted to its environment, and so on. Whereas, for a female specimen, you tend to have more of a story of ‘this is how the species reproduces’.How did the idea of collecting natural-history specimens arise?In a sense, all museums — but certainly natural history museums — have their philosophical origins in the Wunderkammern, the private cabinets of curiosity maintained by aristocrats and natural philosophers in the sixteenth to eighteenth centuries. Some museums today are direct descendants of those collections.During the Enlightenment period, when scientific proof became important, collecting at an institutional level grew. And it went hand in hand with the ‘age of discovery’. Some of those voyages explicitly focused on finding out what resources were out there in the world that could be traded or acquired. And museums were both a place to study those resources — be they animal, vegetable or mineral — and a tool to promote the mission: ‘Look what we’ve got in our newly found colony’.The British colony in southeastern Australia was intended to be founded in what’s now Botany Bay, near Sydney. And the colonists called it Botany Bay — it already had a name, Kamay, in the Indigenous language Dharawal — because, in 1770, on the voyage of Captain James Cook, naturalist Joseph Banks spent weeks collecting plants there. He came back and later said to Parliament, on the basis of what he had found, ‘this is where you should set up a colony’. Those plants are now in the Natural History Museum in London.Specimens at the National Museum of Ireland — Natural History in Dublin.Credit: Lucas Vallecillos/VWPics/Redux/eyevineBanks also wanted to collect the heads of Aboriginal people for his anatomical studies. Isn’t that problematic?It was. Scientists at the time had theorized a racial hierarchy of people across the world, and then sought people’s remains without consent to try to back up this theory. It fed into the eugenics movement starting in the late nineteenth century, which has repercussions today. The desire to categorize people is inherently linked to extraordinary forms of violence and has been used in a pseudoscientific way to justify horrific social policy.How are museums dealing with this legacy of colonialism?There isn’t one museum that I would say is doing colonial history particularly well. But a lot of the research being done in natural history museums is, partly, understanding the true origins of the collections and who really collected them.I bring up nineteenth-century naturalist Alfred Russel Wallace in that context, because he was relatively good at giving credit to people that he was working with. Two Malay teenagers called Ali and Baderoon, in particular, helped him during his eight-year voyage to the Malay Archipelago. Out of the 125,000 specimens that we lazily say were ‘collected by Wallace’, we know he attributed many to other people. But the institutions of science decided to ignore that and give the credit to Wallace.Some museums have begun to repatriate human remains and artefacts. Should every collection be repatriated?The important question is, where does an object have the most meaning? It’s always going to be on a story-by-story basis. Take the thylacine, for example, the extinct Tasmanian tiger (Thylacinus cynocephalus). There are nearly 800 specimens in museums around the world, and there are more in Australia, where they were native, than anywhere else. It is good that all of the thylacines aren’t in Australia. Because we talk about human-driven extinction in museums all around the world, that’s an important story — objects have power by being spread around.Why museums should repatriate fossilsBut there are no specimens of gorillas in museums in any of the gorilla’s home states. If anyone who studies gorillas in Central Africa wants to use museum collections, they have to go to another part of the world, which is not right. If a museum has many gorilla specimens, to repatriate some of them would be a good thing.
Enjoying our latest content?
Login or create an account to continue
Access the most recent journalism from Nature’s award-winning team
Explore the latest features & opinion covering groundbreaking research
Access through your institution
or
Sign in or create an account
Continue with Google
Continue with ORCiD More
Great white sharks (Carcharodon carcharias) are threatened by hunting and poaching.Credit: Education Images/Universal Images Group via GettyFifty years ago today, cinema-goers were scared out of their seats — as well as the ocean, and even swimming pools — with the US release of the film Jaws. Director Steven Spielberg’s creation entertained and terrified in equal measure, and brought about an enduring change in how we perceive sharks.David Shiffman is a marine conservation biologist based in Washington DC, and the author of Why Sharks Matter (2022). He spoke to Nature about how Jaws influenced the public perception of sharks and the importance of protecting marine life.How did Jaws affect the public’s opinion of sharks?There are some people who were afraid to even go into ankle-deep ocean water the summer Jaws came out. My mom says she didn’t want to go swimming at a pool. Around the same time, Star Wars came out, and no one was afraid that the Death Star was going to blow up Earth.A colleague of mine coined the ‘Jaws effect’, which is now used in public-policy literature to explain how fictional portrayals of real-world issues impact what voters want leaders to do about those issues.Has public opinion changed in the 50 years since Jaws was released?More people now love sharks, are aware of shark conservation and want to help sharks than ever before. But there’s still widespread fear, and in some cases it’s quite pseudoscientific in nature.How do these negative perceptions influence shark conservation?Sharks were swimming in the ocean before there were trees on land and before Saturn had its rings. And in the past 50 years, they’ve experienced some of the worst population declines in their 400-million-year-plus history.Unsustainable fishing practices are the largest threat to marine biodiversity. To save the ocean as a whole, including but not limited to sharks, we need to do more sustainable fisheries management. Scientists and environmental advocates in many cases know what we should do, but we’re not the ones in charge. We need to convince policymakers to make changes to conservation policy, and that requires public support.When fear and panic aren’t ruling our emotions, it’s clear that we should be trying to protect these animals, not trying to eradicate them.The film Jaws was released in the United States on 20 June 1975.Credit: Pictorial Press Ltd/Alamy Why is shark conservation so important?Humans depend on the oceans, which provide billions of people with a major source of food and tens of millions with employment and job security. To have a healthy food chain, you need to have a healthy top of the food chain. When we lose predators, the whole system can unravel.How can we change public opinion on sharks?Part of the problem is that most people hear about sharks only in the context of inflammatory media coverage or pseudoscientific nonsense.Yearly, more people are bitten by other people on the New York City subway system than are bitten by sharks in the whole world.The media and popular press definitely have a role to play. It would be great if they would educate about real things or at the very least stop saying nonsense.
Enjoying our latest content?
Login or create an account to continue
Access the most recent journalism from Nature’s award-winning team
Explore the latest features & opinion covering groundbreaking research
Access through your institution
or
Sign in or create an account
Continue with Google
Continue with ORCiD More
