Editorial
General Science

Aiming to Improve on the Reliability of Research Papers for the Good of Scientific Progress

Shinichiro Takezawa

Author and Article Information

Author infoGeneral healthcare Inc.

PublishedMay 30 2014

CitationS. Takezawa (2014) Aiming to Improve on the Reliability of Research Papers for the Good of Scientific Progress. Science Postprint 1(1): e00023. doi: 10.14340/spp.2014.05E0005

Keyword Image-manipulation, Text copying, Image reuse, Figure and table data, Science Postprint

Introduction

Digital images that display bands (*note) showing the results of electrophoresis are often utilized within research papers in the medical field, including in the life sciences. By comparing results against positive or negative controls that act as standards, researchers explain the issues in their papers by showing the existence or lack of such bands, sometimes displaying quantitative results as well.

Therefore, displaying bands is extremely important in the life science research papers. However, image-manipulations have been discovered in certain circumstances, such as when the results of an experiment cannot be duplicated. Recently, the work that claimed to have discovered an astounding type of stem cell 1) dubbed STAP cells is accused of fabricating or falsifying some images.
Investigations into the laboratory notebook of the first author were made, and many people argued its withdrawal. Even related doctoral theses got examined. Co-authors have been covered in the media and forced to hold some press conferences, questioned their responsibility as authorship. Furthermore, the research institute which they belong to have been blamed for their management, which may affect obtaining public funding assistance.

Let us consider then whether the image-manipulation discussed above was a unique case or not. Research paper fabrication consists of a number of different patterns, including copying text from other papers, reusing images, image-manipulation, altering statistical data, completely fabricating sections, and more. What proportion does each method make up of the whole, and how many research papers have been fabricated? What should academic journals do to combat these issues? We should figure out our measures for each of these improper manipulations.

Classification of paper fabrication

1)Text copying

In these cases, the text of other publication is copied, and just some of the keywords are changed into their own unique ones. It might be thought that the researchers from non-English-speaking nations tend to do this because of the language barrier. The copyright infringement is a great concern, but from the scientific point of view, the actual research content is unique, that there won’t be any major issues on science.

2)Image reuse

The reuse of images, which was also an issue in the above case, refers to using, for instance, the image of a cell referred to as “cell A” in one paper, and then using the same image again in another paper, referring to it as “cell B”, which is different from “cell A”. This is a clear form of fabrication, and so either resubmitting factually correct images, or withdrawing the paper, is necessary. These manipulations might occur when the authors cannot meet the peer review requests to show more clear images with the original images. As long as the authors replace images simply because they unfortunately do not have clear ones, it should be possible to obtain the same results by replicating experiments. However, if the images are taken from different experiments to fabricate the scientific fact as they aim to be, it is obvious that the reproductive experiments fail. Such cases present a huge concern not only they lose their confidence but they affect the progress of science itself. For example, if researchers and students consider plans for their next experiment based on such problematic research, they may find themselves puzzled by not being able to replicate the results even after several months, only to reach a conclusion that their results impossible to replicate. Even worse, they might fall into lying over the previous fabrications.

3)Image-manipulation

Image-manipulation was also the method of fabrication in the controversial paper. It seems to have frequently occurred in images used to show the results of electrophoresis. It includes covering base images with other images, matching up electrophoresis images from completely different experiments, and other such falsifying actions. The background of these actions would be the same as seen in cases of the image reuses.

4)Figure and table data

There could also be some cases in which data found in figures and tables are altered. Typical scientific discussion is based on whether a certain value is higher or lower than the control, but if these values are manipulated, researchers can essentially make up any story they wish. Statistics can also be altered in such ways.

If such alterations are simply made as an attempt to make figures and tables look nicer, and attempts to replicate the results of any experiments attain similar results, then there might not be any actual damage on science, but if such alterations are made in an attempt to match previously created stories, it is unlikely such experiment results can be replicated. Such fabrications will negatively impact researchers who use those results for planning their own studies. This is not simply a waste of time, but a waste of research resources, which in turn consumes valuable research funds.

If you simply want to discount the value of research papers by stating that such issues are inevitable, there is nothing left to say, but as Science Postprint, a supplier of academic research papers, we aim to improve the reliability of such academic works and ensure the papers that we publish are highly valuable. Thus, we would like to express our desire to work hard to nip such research paper fabrication in the bud.

Our missions

In terms of the actual activities we are engaging in, we will first focus on the following two points:
1) Conducting survey on the fabrication of research papers
2) Creating a framework for detecting research fabrication

1) Conducting survey on the fabrication of research papers

As explained above, there are a number of patterns that exist within research paper fabrication. A number of points to clarify exist, such as the methods, the amount and so on. We are conducting our investigations in order to clarify these fabrication patterns, including what kinds of methods were used at what level of frequency in past publications, what kinds of researchers have conducted such fabrication, as well as the context.

2) Creating a framework for detecting research fabrication

We are preparing our unique analysis software and accumulating know-how in order to build a framework able to detect research fabrication at the point of submission.

Through these activities we hope to position Science Postprint as a research journal with high scientific reliability. The accumulation of facts by each individual researcher is important for the progress of science, and so we hope our journal is utilized as a place for accumulating scientific facts that do not waver from the truth.

Note: When identifying specific DNA, RNA, or other proteins manipulated in experiments, electrophoresis is conducted on the experiment samples using agar gels, etc., after which proteins, etc., are identified with X-ray film or scanners by using different types of color markers. This results in the signals obtained having a long, thin rectangular shape, and as such these signals are called “bands”.

Reference

  1. Obokata H, Wakayama T, Sasai Y, Kojima K, Vacanti MP, Niwa H, Yamato M, Vacanti CA (2014) Stimulus-triggered fate conversion of somatic cells into pluripotency. Nature 505(7485): pp. 641–647. doi: 10.1038/nature12968.
Editorial is an opinion article on a topical issue. Editorials are not peer reviewed and only receive editing from internal editors. Editorial office may solicit Editorial article from an appropriate member of SPP Editorial Board.