Difference between revisions of "Gnaiger 2019 MitoFit Preprint Arch Editorial"

From Bioblast
Jump to: navigation, search
Line 60: Line 60:
  
 
:::: We need intelligent tools that help to bring validated data into focus . This requires software-supported screening for results that (''1'') have been reproduced by different research groups; (''2'') are backed up with a rigorous quality management, such as pre-publication of time stamped protocols, access to raw data of the complete data set, including calibration procedures and data analysis algorithms; (''3'') are published explicitly by including negative results, which can be compared with corresponding positive findings. Similarly, intelligent tools should become available which put a red flag to publications with conclusions drawn on false statistics. Based upon such and other quality control measures, software detecting plagiarism may be developed further by machine learning into powerful tools for improved filtering of specific research topics, generating databases and context of segmented publications. This will not make scientists redundant, but [[Research |re-searchers]] will have the task to validate such databases as indispensable tools for the advancement of science. A corresponding output metric will be the measurement of the impact of a publication on a database, the value-impact of a database on the knowledge system, and the cultural and socio-economic impact. The publication currency may lose the power to buy the opinion of granting agencies or to serve as an index of scientific prestige. A paper will be traded: many papers are needed to buy a valuable tool or gadget. Even if the individual paper is of little value, many papers together are essential for the whole thing. An industrially produced nail is not worthwhile picking from the street. In context, all publications are potential nails, which need an architect’s plan, quality control for being straight, having the right length, and being made of the proper material. Together with the nails a hammer is needed to build a meaningful structure. Many more nails are needed of sufficient quality and properly inserted to build a structure and hold it together.
 
:::: We need intelligent tools that help to bring validated data into focus . This requires software-supported screening for results that (''1'') have been reproduced by different research groups; (''2'') are backed up with a rigorous quality management, such as pre-publication of time stamped protocols, access to raw data of the complete data set, including calibration procedures and data analysis algorithms; (''3'') are published explicitly by including negative results, which can be compared with corresponding positive findings. Similarly, intelligent tools should become available which put a red flag to publications with conclusions drawn on false statistics. Based upon such and other quality control measures, software detecting plagiarism may be developed further by machine learning into powerful tools for improved filtering of specific research topics, generating databases and context of segmented publications. This will not make scientists redundant, but [[Research |re-searchers]] will have the task to validate such databases as indispensable tools for the advancement of science. A corresponding output metric will be the measurement of the impact of a publication on a database, the value-impact of a database on the knowledge system, and the cultural and socio-economic impact. The publication currency may lose the power to buy the opinion of granting agencies or to serve as an index of scientific prestige. A paper will be traded: many papers are needed to buy a valuable tool or gadget. Even if the individual paper is of little value, many papers together are essential for the whole thing. An industrially produced nail is not worthwhile picking from the street. In context, all publications are potential nails, which need an architect’s plan, quality control for being straight, having the right length, and being made of the proper material. Together with the nails a hammer is needed to build a meaningful structure. Many more nails are needed of sufficient quality and properly inserted to build a structure and hold it together.
 +
 +
== A mission for mitochondrial physiology ==
 +
''In preparation''
  
 
'''Acknowledgements'''
 
'''Acknowledgements'''

Revision as of 15:50, 29 March 2019


MitoFit Preprint Arch         MitoFit Preprint Archives         Editorial mitofit:190002.v2         States and rates mitofit:190001.v6         MitoPedia: Preprints         MitoFit DOI Data Center


Gnaiger 2019 MitoFit Preprint Arch Editorial

Publications in the MiPMap
MitoFit Preprint Archives editorial team (2019) A vision on preprints - for mitochondrial physiology and bioenergetics. MitoFit Preprint Archives (in prep; last update: Gnaiger E, 2019-03-29).

»

MitoFit Preprint Archives editorial team (2019) MitoFit Preprint Archives

Abstract: A manuscript prepared for traditional journal publication on ‘Mitochondrial states and rates’ is the first preprint published in ‘MitoFit Preprint Archives’ (Gnaiger et al 2019). It actually triggered the initiation of a preprint server for mitochondrial physiology and bioenergetics. In this editorial we tell the story behind starting another preprint server, explain the rationale of integrating it within the Bioblast and MitoPedia websites, and develop a vision of science communication beyond traditional journal and preprint publication from the perspectives of unsustainable exponential growth of increasingly fragmented literature, challenges of priority and reproducibility, and the struggle with relevant data, irrelevant information and revelation of knowledge.


A brief history of preprints

What would today be considered as a preprint server was initiated in 1961 at the National Institutions of Health (USA) as ‘Information Exchange Groups’. These were shut down in 1967 due to the resistance of publishers rejecting articles that were made available as preprints (Cobb 2017). It took 30 more years to start the modern concept of preprints, when Paul Ginsparg initiated arXiv in 1991 for physics and mathematics. arXiv is maintained and operated by Cornell University (USA). The field ‘quantitative biology’ was added to arXiv in 2003. Ten years later, bioRxiv was launched in 2013 by Cold Spring Harbor Laboratory Press in New York as a ‘free online archive and distribution service for unpublished preprints in the life sciences’, with the intention of complementing arXiv. In the same year, PeerJ Preprints was launched as the 'pre-print' area of the Open Access journal PeerJ, accepting submissions in the same subject areas as PeerJ (biological, medical and environmental sciences) and PeerJ Computer Science. Similarly, Preprints started in 2015 associated with the Open Access journals operated by MDPI, based in Basel (Switzerland).
Many discussions on preprints focus on quality standards, peer review, and time delays of publication (Vale 2015). An interesting open peer review model is presented by the Open Research publishing platform F1000Research: Submitted manuscripts are posted to the journal’s website immediately with a DOI and without editorial bias. Subsequently, a transparent peer review is initiated with non-anonymous experts and ‘driven by the authors who must suggest the reviewers and who decide when and how to address any criticisms raised by the reviewers’. The reviewers' reports, reader comments, and the authors’ responses are published, and new versions of the manuscript include explanations of the changes. Completion of the review process entails upgrading of the manuscript as a publication indexed in PubMed and other biobliographic databases. Peer review may be stopped, while the article with a DOI cannot be removed. Then it is equivalent to a preprint, open for submission to another journal for peer review and publication.


From a MitoEAGLE preprint to the MitoFit preprint server

Is a new preprint server needed? How can the attempt to offer a new preprint server be justified? These are difficult but decisive questions to be answered when initiating MitoFit Preprint Archives (mtFPA).
First, launching a preprint server is different from starting a new journal: like all preprint servers, mtFPA does not compete with traditional or Open Access journals, is non-profit and thus contrary to predatory publishers, and does not enter the arena of conventional impact metrics. Preprints are not peer-reviewed, but reflect the time-stamped state of the author's work presented to the scientific community. Differences in scientific opinion of authors, referees and editors will not lead to any delay, let alone rejection of a preprint manuscript. 'Journals then may be incentivized to look more toward quality than speed and seek to publish the definitive work that will stand the test of time' (Vale 2015).
Second, mtFPA is not fundamentally different from established preprint servers. The amazing success of arXiv can be gauged from its mere size and, importantly, gains from the large number of articles (roughly 757,000 from 1991 to 2012) which helped machine-learning algorithms to detect and analyze patterns of copying published text; thus articles with a high duplication score are automatically labelled by a plagiarism filter as an alert to readers and authors alike (Citron, Ginsparg 2015). At the start of mtFPA, the very small size sticks out as a specific advantage, when we can expect a gradually growing number of manuscript submissions in the fairly well defined area of mitochondrial physiology and bioenergetics
(1) In contrast to physics and mathematics, where publication in arXiv has become a standard, preprints remain relatively unpopular in biomedical sciences. A specific mitochondrial physiology (MiP) preprint server can encourage more authors to engage in preprint publications within the MiP network.


Questionnaire Graph.png

Figure 1. The ‘States and rates questionnaire’: Three questions on preprints were circulated on 2019-02-12 to the 530 coauthors of Version 1 of the preprint Mitochondrial respiratory states and rates. 179 responded by 2019-03-15. 38 coauthors were unfamiliar with the concept of preprints. Among those, 20 responded positively to publish in MitoFit Preprint Archives, whereas only 8 were against preprint publication. Note that among 'alternative preprint servers' a majority of 67% (120) selected 'no other suggestions', 29% (52) selected bioRxiv, and a few websites were listed which actually do not qualify as preprint servers (ASAPbio, PrePubMed). Nature Precedings ceased accepting new submissions in 2012. In summary, 'MitoFit Preprint Arch' is a positive contribution to make preprints more popular in mitochondrial physiology.

(2) A growing number of manuscripts may receive special attention by a comparatively large number of readers. Readers can become voluntary members of the 'International editorial board', and join as team players helping to increase reproducibility and traceability. Readers and editors may provide critical and constructive suggestions even after a preprint version has been published, in the spirit of open communication in contrast to anonymous peer review. By mutual agreement between authors, editors and contributing readers, relevant sections of editorial correspondence may be published as a supplement added to follow-up preprint versions, or as a joint editorial with an independent DOI.
COST Action MitoEAGLE
Our analysis of the ‘States and rates questionnaire’ (Fig. 1) provides visible support for launching the preprint server for mitochondrial physiology. MitoFit Preprint Archives (mtFPA) is as such a product of the COST Action CA15203 MitoEAGLE (www.mitoeagle.org). mtFPA is expected to grow with and beyond the MitoEAGLE consortium under the umbrella of the international Mitochondrial Physiology Society (www.mitophysiology.org).


Bioblast, MitoPedia, and Gentle Science

Bioblast wiki
The website Bioblast (www.bioblast.at) was brought to life in 2010 by Oroboros Instruments as a glossary and index for high-resolution respirometry. Bioblast was presented at the Mitochondrial Physiology Conference 2010 (Renner-Sattler, Gnaiger 2010), including the ‘MitoPedia’ as a Wiki, driven by scientists with an Open Access concept. Without being familiar with the concept of preprints, we considered the section on ‘Publications’ to provide “a portal for sharing, disseminating and commenting relevant literature in mitochondrial physiology, with context–related ‘filters’ for references. Bioblast allows the evolution of a scientific publication – providing space for open discussions and extensions of an otherwise static paper”.
MitoPedia
MitoPedia is an ongoing effort to establish a high-resolution terminology, as a dynamic tool for summarizing definitions of terms, symbols and abbreviations. Catalytically working as an Information synthase (from ‘ATP synthase congregation’ by Odra Noel), MitoPedia supports the decentralized evolution of a glossary of scientific terms, for developing a consistent nomenclature in the growing field of mitochondrial physiology.


MitoFit pdf
The MitoPedia section on Preprints (MitoPedia: Preprints) was added in conjunction with the launch of MitoFit Preprint Archives, emphasizing the spirit of Gentle Science in our aim to connect and collaborate with preprint ambassadors (see ASAPbiol). The Bioblast Wiki provides a platform for Gentle Science in the spirit of Scientific Social Responsibility, and MitoFit Preprint Archives constitutes a timely fit in this context.


An industry of scientific publication

Science is progressively turning into an industry with unchecked floods of publications in the business channels of scientific journals. A recent whirl of commercial Open Access journals and the non-profit wave of accelerated preprint publication leads to further swelling of the stream. Many labs used to celebrate a new team publication with a toast. The number of scientific publications posted per day is growing faster than the motivation to throw another party and overrides the capacity to turn scientific innovation into knowledge. Publications as a currency of scientific output are subject to an increasing inflation rate, just as economic inflation is driven by excessive money supply. Publications should be seen as a currency, in contrast to scientific output in terms of goods and services provided by research. Reproducible results, reliable databases, relevant information, meaningful knowledge obtained by putting information into context lead to scientific output, such as a diagnostic test for early Alzheimer diagnosis and treatment of dement patients, or effective preservation of environmental resources.

Distinguishing scientific goods and services from inflation of the publication currency

A scientist lists her or his publication record to obtain an academic degree and position or to support a grant application. Publications in research may be compared to a bank account in the world of investors. Evaluation of a currency depends on how much money can buy. Of course, the Syrian pound (SYP) is not rated equal to the British pound (GBP); a paper in a local journal is not rated equal to a publication in a globally distributed international journal. Academic productivity is measured by widely applied publication metrics, such as the journal-impact factor or h-index (Carpenter et al 2014). The focus on publication currencies, however, detracts from the question, which goods the currency can buy. Irreproducible results published in traditional journals or preprints have a negative impact on society, if any. Is the number of patents a better metric for innovation (Silver 2012) compared to publications? What is the value-impact on society of a high journal-impact publication, versus the value of a preprint that may be influential irrespective of conventional publication metrics? Good measures of scientific impact on society may not yet be available. A value-impact factor is difficult to define. But this should not detract us from searching for intelligent solutions, their optimization and implementation. Both, traditional journal publications and preprints contribute towards progress in improving the scientific goods and services, even at an increased inflation rate and decline of the currency value of each unit of scientific publication.

Worthless but useful – are publications comparable to nails?

PubMed lists 8, 13, 26, and 46 publications per day in 1988, 1998, 2018, and 2018, when searching for ‘mitochondr*’. How many of those did I miss without being aware of their possible relevance for my research or the manuscript currently in preparation for submission? Is another ‘paper’ added to the masses still considered as a potential building block of human culture? Reviews have been a help in science and teaching. But a review published today will be outdated tomorrow. So what’s the point of reading or even citing it?
We need intelligent tools that help to bring validated data into focus . This requires software-supported screening for results that (1) have been reproduced by different research groups; (2) are backed up with a rigorous quality management, such as pre-publication of time stamped protocols, access to raw data of the complete data set, including calibration procedures and data analysis algorithms; (3) are published explicitly by including negative results, which can be compared with corresponding positive findings. Similarly, intelligent tools should become available which put a red flag to publications with conclusions drawn on false statistics. Based upon such and other quality control measures, software detecting plagiarism may be developed further by machine learning into powerful tools for improved filtering of specific research topics, generating databases and context of segmented publications. This will not make scientists redundant, but re-searchers will have the task to validate such databases as indispensable tools for the advancement of science. A corresponding output metric will be the measurement of the impact of a publication on a database, the value-impact of a database on the knowledge system, and the cultural and socio-economic impact. The publication currency may lose the power to buy the opinion of granting agencies or to serve as an index of scientific prestige. A paper will be traded: many papers are needed to buy a valuable tool or gadget. Even if the individual paper is of little value, many papers together are essential for the whole thing. An industrially produced nail is not worthwhile picking from the street. In context, all publications are potential nails, which need an architect’s plan, quality control for being straight, having the right length, and being made of the proper material. Together with the nails a hammer is needed to build a meaningful structure. Many more nails are needed of sufficient quality and properly inserted to build a structure and hold it together.

A mission for mitochondrial physiology

In preparation

Acknowledgements

Acknowledgements

I thank Marija Beno for summarizing the ‘States and rates questionnaire’ and preparing Figure 1. Contribution to COST Action CA15203 MitoEAGLE, supported by COST (European Cooperation in Science and Technology).

References

  1. ASAPbio: Accelerating Science and Publication in biology - http://www.mitofit.org/index.php/ASAPbio
  2. Callaway E (2013) Preprints come to life. Nature 503:180. - »Bioblast link«
  3. Carpenter CR, Cone DC, Sarli CC (2014) Using publication metrics to highlight academic productivity and research impact. Acad Emerg Med 21:1160-72. - »Bioblast link«
  4. Citron DT, Ginsparg P (2015) Patterns of text reuse in a scientific corpus. Proc Natl Acad Sci U S A 112:25-30. - »Bioblast link«
  5. Cobb M (2017) The prehistory of biology preprints: a forgotten experiment from the 1960s. PeerJ Preprints 5:e3174v1 https://doi.org/10.7287/peerj.preprints.3174v1.
  6. Gnaiger E, Aasander Frostner E, Abdul Karim N, Abumrad NA et al (2019) Mitochondrial respiratory states and rates. MitoFit Preprint Arch doi:10.26124/mitofit:190001.v2. - »Bioblast link«
  7. MitoPedia: Preprints - http://www.mitofit.org/index.php/MitoPedia:_Preprints
  8. Renner-Sattler K, Gnaiger E, eds (2010) Mitochondrial physiology. The many functions of the organism in our cells. Mitochondr Physiol Network 15.6:128 pp. ISBN 978-3-9502399-4-2. - »Bioblast link«
  9. Silver N (2012) The signal and the noise. The art and science of prediction. Penguin Press:534 pp. - »Bioblast link«
  10. Vale RD (2015) Accelerating scientific publication in biology. Proc Natl Acad Sci U S A 112:13439-46. - »Bioblast link«


Labels: