Nature 's roundup of the papers and issues gaining traction on social media.

Based on data from Altmetric.com. Altmetric is supported by Macmillan Science and Education, which owns Nature Publishing Group.

Thanks to an analysis in the British Medical Journal (BMJ), social media is hosting the latest round of the debate over medical studies involving animals. Researchers have also used Twitter to pass around a blistering critique of academic publishing — a paper that had observers buzzing over its fate as much as its content.

Writing in the BMJ, Yale University epidemiologist Michael Bracken and UK medical sociologist Pandora Pound argued that too many animal trials investigating medical treatments are poorly designed, and called for better use of systematic reviews to maximize their benefit. Lenny Verkooijen, a clinical epidemiologist at University Medical Center Utrecht in the Netherlands, tweeted that there is “insufficient systematic evidence for the clinical benefits of animal research”. But in a letter to the journal, pharmacologist Fernando Martins do Vale at the University of Lisbon noted that animal research has benefited medicine and has led to “seminal discoveries in the field of physiology, biochemistry, pharmacology and genetics”.

Bracken and Pound cite several examples of ultimately futile animal research. They write that decades of stroke studies using animal models have yet to yield a single treatment that is useful for humans. Likewise, they note, not one of the more than 100 drugs that have been tested in a mouse model of amyotrophic lateral sclerosis has proved to be beneficial after further experiments. Part of the problem, they argue, is basic biology: the gulf between lab animals and humans is often too wide for meaningful extrapolation. But they also see basic shortcomings in how animal studies are carried out. They cite an analysis of 271 animal studies conducted between 1999 and 2005, which found that only 12% included randomization for treatments and controls. Systematic reviews — which are relatively uncommon in animal research — have pointed to widespread bias in reporting and publishing of results, they write.

In an e-mail, Bracken clarifies that he isn't calling for any sort of ban or moratorium on animal research. “We are saying that badly designed ... animal research, which is now widely documented to be commonplace, is unethical.” Not only do such trials fail to benefit humans, he says, but they can also actively harm people who are enrolled in clinical trials of drugs or other therapies.

Responding to the article, several letters to the editor published online by the BMJ showed broad agreement with Bracken and Pound's position, but some respondents felt that the benefits of well-conducted animal research had been understated. Martins do Vale also noted in his letter that “many discoveries that are fundamental pillars of the modern medicine resulted from experiments ethically intolerable on humans and only possible with the animal experimentation.” Physicians Andrew Whitelaw and Marianne Thoresen, who specialize in neonatal neuroscience at the University of Bristol, UK, write that two effective treatments for premature infants — prenatal corticosteroids and postnatal surfactant replacement — were developed using animal models.

On Twitter, many researchers praised the paper and concurred with its conclusions, but a full-throated defence of animal research was not evident. Although Bracken didn't follow the Twitter discussion, he says that the overall reaction to the BMJ paper has been “overwhelmingly in agreement”. He notes a blog post from Understanding Animal Research, a UK group that promotes the use of laboratory animals, that conceded that animal research hasn't always been sufficiently rigorous. In the post, policy and communications officer Liz Harley wrote that “many of the authors' suggestions for improvement are already being addressed”. But she also noted that “it is disingenuous of the authors to imply that the sector is either unaware of, or worse actively ignoring these problems.”

In an e-mail, Wendy Jarrett, chief executive of Understanding Animal Research, says that Bracken and Pound are “grossly exaggerating the limitations of animal models”. Carefully conducted animal studies, she argues, can greatly advance medical knowledge. “Practically all of the human and veterinary medicines we have available today were developed and tested using animal models,” she says. “We are aware that humans are not mice, but using the correct animal for the correct purpose has been critical in the history of medical progress.”

Pound, P. & Bracken, M. B. Br. Med. J. 348, g3387 (2014)

Another paper making the rounds on social media seemingly fell victim to a self-fulfilling prophecy. The article, a critique of the profit motives and business models of the academic publishing industry that appeared in the journal Prometheus: Critical Studies in Innovation, ran into interference from the journal's publisher, Taylor & Francis. The paper — by researchers from the School of Management at the University of Leicester, UK — says that the publishing industry, in its fight to protect large profit margins, “is as resistant to change as the music publishing industry was”. Just as that industry survived the invention of downloads and file sharing, the paper posits, academic publishing could thrive with more open access. Senior managers at Taylor & Francis initially refused to run the paper — entitled 'Publisher be Damned! From Price Gouging to the Open Road' — unless large sections were deleted. Publication was delayed for months. When the piece finally did come out mostly intact, the PDF version carried a disclaimer including the phrase “the accuracy of the article should not be relied upon”.

The social-media reaction was one-sided. Twitter users expressed frustration with academic publishing in general and the incident with Taylor & Francis in particular. Jon Tennant, a PhD student in palaeontology at Imperial College London, tweeted that the paper shows “the general damage publishers cause”. Zen Faulkes, an invertebrate neuroethologist at the University of Texas-Pan American, discussed the actions of Taylor & Francis on his Neurodojo blog: “It is rare to have such obvious meddling over such an obvious conflict of interest.”

Stuart Macdonald, general editor of Prometheus, who studies the influence of information and innovation on policy at Aalto University in Helsinki, said in an interview that he was “astonished” by the publisher's actions. “The industry clearly does not want to discuss the industry,” he says. “The issue that's of most importance in this miserable episode is whether academic publishers appreciate that they must not intrude on editorial independence.”

Macdonald says that he and other editors initially threatened to quit over the matter, but he is instead waiting for a public apology that he says the publisher promised him. As of 16 June, Taylor & Francis had not issued an apology.

David Green, global journals publishing director for Taylor & Francis, says: “We are still constructively working through these matters ... but I'm afraid I can't comment further at this time.”

Harvie, D., Lightfoot, G., Lilley, S. & Weir, K. Prometheus 31, 229–239 (2014)