Scientific journals, manipulated data and science activists

Scientific journals, manipulated data and science activists

For some years now, some have been questioning the scientific publishing world. But who are they and what are they talking about?

Today we wanted to explain in detail how the paper publishing system works and who are some of the people who are revolutionizing it

Before we start talking about some of these scientists, it is important to understand how te scientific publishing works. When a scientist wants to publish his results, the process that is followed can be summarized in three steps:

1. Sending the manuscript to the journal: The scientist submits his work to a journal, with the commitment that he has not sent it to any other journal. The work is reviewed by the journal’s editors, who decide whether the work can move on to the next phase.

2. Peer review: At this point, experts in the field review the work and tell the editor if it is good and should be published or rejected. These expert scientists have nothing to do with the journal and charge nothing to do this work. It should also be noted that the authors of the work do not know who is reviewing their work, but the reviewers do know the names of the authors. Editors usually send the article to different reviewers and compare their views to make the decision about the publication.

3. Pay to publish: If the journal finally decides to publish the article, it charges the authors to do so, supposedly to cover the costs of publication. The publication price depends on the magazine, its “prestige” and its impact factor, but can range from 1000€ to 4000€. Some magazines also give the possibility of publishing in open access, but obviously this raises the cost.

 

Alexandra Elbakyan and Sci-Hub

Today nearly all scientists use Sci-Hub. What is it, how was it created and why?

Alexandra Elbakyan has a degree in computer science and has been working in research for years. In 2009, when she was doing her thesis, she ran into the problem that she constantly needed to pay to have access to scientific articles. The price of a paper was around $ 30. Universities and centers usually pay for a subscription to journals so that people who access them from their networks can read them, but depending on what you pay for, you will have access to more or fewer articles. Alexandra Elbakyan, who could not afford this cost per paper, set out to investigate how to hack them in order to obtain them. Two years later Sci-Hub was born, a website that allows you to download millions of articles for free.

Screenshot of the Sci-Hub homepage

Obviously, this didn’t please scientific journals, Elsevier magazine even filed a lawsuit against her. She didn’t win the trial (although she did not pay the fine imposed, as she lives in Russia, outside the United States’ jurisdiction) and was forced to close Sci-Hub, which it was quickly reopened with another domain.

This story leads us to ask, why did Sci-Hub make magazines so angry but not the researchers?

The reason why researchers are not angry with Elbakyan, but on the contrary, some call her the “the Robin Hood of Science” is easy to understand knowing the editorial workings. Scientists generate the content of journals, they have to pay to publish and to pay back to read what others have published. That’s why probably Elbakyan had no personal detractors and those who have opposed Sci-Hub are the magazines seeing their profits endangered. In fact, more and more Sci-Hub papers are being downloaded every day.

Downloads per day on Sci-Hub for months with access logs. Himmelstein, D. S. et al. eLife 2018;7:e32822
Downloads per day on Sci-Hub for months with access logs. Himmelstein, D. S. et al. (2018) 

In the last years, open access journals have also been created. Sci-Hub has become so popular that a Google Chrome extension has just appeared allowing you to directly search for the article in Sci-Hub and download it for free. It also operates on Telegram for when the domains it uses are closed.

 

Heather Piwowar, Jason Priem, Cristhian Parra and Unpaywall

Unpaywall logo
Unpaywall logo. Heather Piwowar and Jason Priem / CC BY https://creativecommons.org/licenses/by/4.0

In the same line, although working slightly different, there is Unpaywall. These three computer scientists spent one night, after a conference, working on this project. They developed a software that tracks on the network for locations where papers are available for free. That is, if we find a paper that apparently has to be paid for reading, Unpaywall looks if it exists in open access elsewhere and redirects you there. Unpaywall also works through a browser extension.

 

Beyond open access; data manipulation

Since it is becoming easier to have access to all papers, everything that is being published can be now more reviewed and monitored by the scientists who read it. In the last years, numerous scandals over data manipulation have come to light. To talk about it, we think it is appropriate to explain who are those who have become famous for denouncing this kind of corruption.

Leonid Schneider and the complaints to the scientific world

Alexandra Elbakyan questioned the model of the scientific journals, but Leonid Schneider, a science journalist and researcher, has directly called into doubt the veracity of scientific publications. Schneider is the author of the “For Better Science” platform and he denounces that the scientific world is only giving a voice to the elites. Briefly, he says that some scientist receive certain privileges, thus creating a reputation that makes unquestionable everything they do. Participating in this process, there would be the scientific journals but also politicians. With “For Better Science” he tries to create a platform that allows to criticize the others’ work, fraud or ethical non-compliance, regardless of your personal status. He also has a Twitter account, @schneiderleonid, in which he exposes and criticizes what he believes that is corruption, favoritism or malpractice.

Leonid Schenider is known for exposing big scandals in the scientific world and he has even go on trial against some researchers. Since he is a controversial person, he has been interviewed several times (available on his website).

Which are Leonid Schneider’s sources? Let’s talk about PubPeer

He says that everything that he investigates comes from material that people send to him, but also papers that are posted on PubPeer. PubPeer is one of the several existing platforms that allows people to do the famous “peer reviews” that we talked about, once the papers are published. The website is defined as “the online Journal Club”. They promote that the entire scientific community can be actively involved in the assessment and judgement of the others’ work. Such platforms have led to many other scientists to develop results verification tasks.

 

Elisabeth Bik and her work looking for data manipulation

While Leonid Schneider is, perhaps, taking a more activist stance or he is fighting against the elites, Elisabeth Bik is one of the most visible faces in the verification of data published in scientific papers. She is a microbiologist and she has gone from the research to being a science consultant. She is mainly engaged in looking for fake results such as manipulated images using Photoshop or other image editors. She has a personal blog, Science Integrity Digest, were she posts some of these cases.

In her Twitter account, @MicrobiomDigest, she also posts some cases as it was a game. She uploads images from papers and asks to her followers if they are able to find the image retouching or mistakes.

Preprint repositories: arXiv, medRxiv, bioRxiv and similar

Secreenshot of the bioRxiv homepage

Knowing all these things, it seems that science is becoming more open to everyone. There are other contributors to this factor as the preprints repositories. These are papers that have not been yet peer reviewed. These repositories allow the readers to comment the papers and therefore allow the entire scientific community to review them, as in PubPeer, but with the difference that papers have not been published. bioRxiv is focused on the research of biological sciences, arXiv was the first one created, and papers were mainly from the field of physics and medRxiv is aimed at the studies of health sciences and medicine.

These lasts months, during the COVID-19 pandemic, bioRxiv has been one of the main sources of the disseminated results. The main advantage, as it has been said, is that papers can be viewed and read for everybody before the review and publication. At the same time, this could be an inconvenient, since these papers and results that have not been verified by anybody can be rapidly diffused. Thus, one has to make the effort to read them with calm and to find out if what it is being exposed can be contradictory or even contain fake data.

 

This raises the following question: are we capable to evaluate in a personal manner all the scientific content?

It is hard, at a personal level, that a researcher can perform his work and, at the same time, he can be able to review everything that he reads, constantly. Luckily, one of the great advances of the last years is the deep learning, which allows that computer systems can perform complex automatic analyzes. Thanks to these tools, scite has been recently created; a software that analyzes in an automatic way how a paper is cited by others, so that it creates a classification of other papers that support or refuse the evidence. Thus, it is a tool that, although it is not able to directly find out if there is fake data, it allow us to see the results in a larger context, and so provides information to know if the results being exposed are true.

Screenshot of scite homepage

In short, easy access to scientific content has revolutionized a system that, according to many, till now only benefited a few. But in addition, this allows that the information is available for everyone, leading to stricter reviews. Many claim that, in the same way that scite has been created, in the next years other softwares are going to appear and that they will directly detect falsified images, so it will be increasingly easier to detect bad practices.

Speaking on behalf of Micronautes, we firmly believe that knowledge must be public and accessible to everyone. We also hope that this surveillance to which we are now subjected, will alert those who are willing to falsify their results for the purpose of publishing. Scientific fraud mainly harms scientists, delays the others’ work and provides tools to those who question the veracity of science in order to attack it.

 

References

  • Featured Image: 2020. [image] Available at: <https://upload.wikimedia.org/wikipedia/commons/7/7e/Scientific_citations.png> [Accessed 25 May 2020].
  • Es.wikipedia.org. 2020. Alexandra Elbakyan. [online] Available at: <https://es.wikipedia.org/wiki/Alexandra_Elbakyan> [Accessed 21 May 2020].
  • Aluffi, G., 2020. Entrevista | “Es Cierto: Robo A Los Editores Para Dárselo A Los Científicos”. [online] EL PAÍS. Available at: <https://elpais.com/elpais/2017/02/06/ciencia/1486399819_243966.html> [Accessed 21 May 2020].
  • Mingarro, I., 2020. Editoriales Científicas: De La Necesidad… Negocio. [online] Sebbm.es. Available at: <https://www.sebbm.es/web/es/divulgacion/rincon-profesor-ciencias/articulos-divulgacion-cientifica/2469-editoriales-cientificas-de-la-necesidad-negocio> [Accessed 21 May 2020].
  • Monbiot, G., 2020. La Estafa De Las Revistas Científicas Se Acerca A Su Fin. [online] eldiario.es. Available at: <https://www.eldiario.es/theguardian/estafa-revistas-cientificas-acerca-fin_0_814368712.html> [Accessed 21 May 2020].
  • Villatoro, F., 2020. El Coste Real De Las Publicaciones Científicas, Ese Gran Desconocido – La Ciencia De La Mula Francis. [online] La Ciencia de la Mula Francis. Available at: <https://francis.naukas.com/2013/03/28/el-coste-real-de-las-publicaciones-cientificas-ese-gran-desconocido/> [Accessed 21 May 2020].
  • Else, H. 2018. How Unpaywall is transforming open science. Nature 560, 290-291. doi: 10.1038/d41586-018-05968-3

Leave a Reply

Your email address will not be published. Required fields are marked *