email enquiry to BioinfoTools back to BioinfoTools home page back to BioinfoTools home page

Bioinformatics tools

Software development

Training

Consulting

 

Overview

FAQ

Ruminations

Products

Resources

Contact us

Legal

An 'executive summary' of BioinfoTools aims

Answers to frequently asked questions about BioinfoTools

Read Dr. Jacobs' occasional opinion piece

Learn what BioinfoTools is developing

Links, tutorials & free software

Getting in touch with BioinfoTools

Copyrights & liabilities


Quick jump to: BioinfoTools Home page  >> Ruminations (this page)
 

Ruminations

revised: 12-Dec-02
Copyright  © 2001, 2002 BioinfoTools.com
 
Ruminate /'ru:mineit/ v. [L ruminatus pa. ppl stem of ruminari, -are, f. as RUMEN: see -ATE.] 1. v.t. Turn over in the mind; mediate deeply upon. MI6 2. v.i. Chew the cud. MI6 [etc.]

Taken from The New Shorter Oxford English Dictionary (1993).

Current column

#2 Credits, Dis-credits and mis-credits.

Past Ruminations

#1 The mythology of bioinformatics.



Articles at other sites

American Museum of Natural History "The Genomic Revolution" exhibit

Bioinformatics (BusinessWeek)

Interview with author's of O'Reilly's 'Developing Bioinformatics Computer Skills ' book.
 

This occasional column is inspired by the "Ask Tim" column of publisher Tim O'Reilly of O'Reilly books fame. Whereas Tim's column is primarily in response to letters he has received, items here may be inspired by nothing much at all. This column will be occasional -- that is, whenever I (Grant Jacobs) find time and inspiration (whatever that is) to put down some of the stray thoughts in my mind. My ruminations are unlikely to bear the full weight of the definition to the left -- they'll be more light-hearted. But hopefully, they'll raise issues worth mulling over.

You're welcome to write in. Please don't be offended if you don't get a reply promptly, or even no reply at all, as I am likely to occupied with other matters. But I will read everything posted. Posts that raise interesting issues may get a reply via this column. Posts should be sent to ruminate@bioinfotools.com.

Topics will span a wide range of areas, such as computational biology, ethical issues in research, computing, computer hardware, business practices and possibly other areas of science (e.g.. cognitive neuroscience). Don't feel shy if your topic or question seems unusual or irregular. If anything, topics which are different enough to jar me out of my train of thought and make me say "hey, now thats a thought!" are more likely to receive attention.




Current Rumination

Rumination #2.
Credits, Dis-credits and mis-credits.

Copyright Grant Jacobs, BioinfoTools (2002-)


In a recent issue of Nature (21st February), Peter Lawrence, a senior scientist at my Ph.D. training grounds (the MRC Laboratory of Molecular Biology, Cambridge, UK) wrote what a related editorial titled "Thoughts on (dis)credits". (Peter's article is entitled "Rank injustice".) I'll like to recap on this issue and suggest a possible solution.

The basic issue appears to be that of allocating credit for work on publications. Peter points out that the existing system, at least in molecular biology, is practically set up to advance those already in positions of high rank at the expense of those who work underneath them should they choose to use the opportunities available.

Or, as Peter opened his article: "What has rank to do with the process of creative science? Very little. What has rank to do with the politics of science and the allocation of credit for discoveries? Almost everything." Of the many issues covered in Peter's article and the associated editorial, most seem to largely boil down to the issue of senior authors taking credit for work they have not had an active input in on.

Peter Lawrence is, of course, referring to "criminal" credit - stealing credit from others. Here I'd like to tackle the more mundane cases in the hope that solving this would eventually reduce the incidence of the more "celebrated" cases. While I deplore dishonesty in science (I have been the recipient of it myself), I doubt any system will truly get rid of those who wantonly garner credit, just as plagiarism is essentially impossible to stamp out. We can merely hope to reduce the scale of the problem and provide practical schemes which encourage honesty.

My suggestion would be for journals to present two rows of credit lines: one for the authors who actually did the research work and a second recording the group leader, etc., perhaps including the head of the department in question. Group leaders may not have (much) contributed to the actual science, but run the laboratory, are likely to have obtained the grants for the research, and so on. The argument over credit should not be that these people deserve no credit, but that they deserve a different kind of credit. After all administration does take a good deal of effort and deserves credit. However, what is needed is a mechanism so this different form of credit can be distinguished from scientific credit.

In journals, it would seem to me at least three sorts of credits could be imagined: (1) The senior author(s) of the actual research, (2) the remaining authors in some basic order of merit and, separately, (3) the administrative leader(s). To do this each journal ought to: (a) identify where the senior researcher is expected to be in the author list; (b) if there is any (rough) priority order, state what it is; (c) and finally separately identify the administrator the work fell under. This information can be held under the usual guidelines for the journal, with the papers themselves merely presenting the appropriate names. For example, we might have:

    Authors: A.P.I. Brainy, R.A.D. Postdoc, ..., L.I.L. Student
    Administrators: B.I.G. Shot

Somewhere in the journal guidelines would be a statement along the lines of: "The senior contributor is to be placed first in the authorship list, with the resulting authors in approximate order of decreasing merit. Administrators ought only be placed on the authorship if they genuinely contributed to the science. Administration and funding support should not be construed as scientific contribution." Different journals will, of course, have differing schemes, but the essential thing is that a mechanism of presenting the different credits be present and understandable to the reader.

Likewise, bibliographical databases (eg. Medline via PubMed) need to record these three items. A survey of their subscribing journals would yield the information at comparatively little cost. Obviously, there would be issues with implementing the new keywords, etc., but should be more a matter of basic motivation than anything else.


The remainder of the article explores individual cases in a little more detail.

In molecular biology (and bioinformatics) there is a strange dichotomy, whereby the last author is generally assumed to be both the group leader and senior contributing author with the first author being the actual person who has done the bulk of the day-to-day research. This leads to the situation whereby the last author captures senior credit regardless of whether they contributed to the conception or execution of the research.

While there are certainly those who exploit this as Peter Lawrence's article illustrates, many are simply flies stuck in a rather sticky web. Giving a separate credit line would allow at least those innocently trapped to free themselves and still retain appropriate credit. Or, for the less innocent, it would nip the habit in the bud before like petty criminals they move on to more serious crimes.

It is a rare scientist who can afford to miss credit. Certainly those in the MRC Laboratory of Molecular Biology (LMB) can hold their heads high with regard to not inappropriately taking credit from younger colleagues. My own single author EMBO J. paper as a student is a direct reflection of this attitude at the MRC LMB. In another article I may explore what I believe are the elements behind the success of the LMB (it has gathered 9 Nobel prizes to date) and why the scientists there hold themselves to high ethical standards. An important point to note is that you can both succeed and hold yourself to high standards.

Imagine a post-doc in a large laboratory headed by a well-known senior scientist. The post-doc conceives a brilliant line of research and obtains the use of a student to do most of the actual work. It is difficult for the post-doc to receive honest credit for his work, unless the senior scientist concurs not to place their name on the paper as senior contributor. Naturally, the senior scientist founded the laboratory, administrates it and probably obtained the grant which the post-doc and student is employed on. The senior scientist certainly deserves credit for creating an excellent research environment and perhaps mentorship, but not at expense of poaching the credit for the research itself.

Likewise, a senior researcher may appreciate that their appearance as an author may imply that they conceived of the work and that their student or post-doc "merely" executed it, robbing them of their rightful credit. This is particularly true with especially prominent researchers and to the credit of some, they have left their name off papers of the staff they support where appropriate. An awkward compromise here is that it is sadly true that the presence of the respected scientist's name would probably aid publication in a high-quality journal, claims of assessing the science on its own merits notwithstanding.

A related cultural issue is senior scientists honestly admitting their role in the science "industry". Some older scientists are no longer productive "at the coal-face", perhaps because their administrative duties prevents time to follow the literature in sufficient depth or because their interest has waned. But because their name is known and they understand well the process of science, particularly at their institution, they are useful administrators. Being purely an administrator is often viewed as being a negative thing - losing one's scientific acumen as it were. Some of these individuals appear, to me, to be reluctant to openly acknowledge their contribution is no longer on the actual research under their wings. In order to attract more funds and to continue to write grants with their name on them, they really do need their name on the publications arising, but they cannot rightfully claim research credit. An administrative credit would be appropriate.

An obvious remaining issue to the suggestion offered here is credit for conceptual research design as opposed to its execution. It is very common for the research to be conceived, planned and coordinated by an more experienced researcher, with younger staff taking on the actual research under their direction. Again we have different types of credit. In this case, priority is generally granted to the conception of the research, even if the younger staff (including technical staff) have had to devise some original experimental design or technique to achieve this research. I wonder how many important discoveries are in a practical sense dependent on some "lowly" technician's good sense to devise a method "to make the boss' idea work"? I can only recommend that researchers take the time to understand what their staff's contribution really are. Some may feel unable to speak up for fear of job security, shyness, etc.

A colleague of the author, Dr. Ines Carrin, pointed out that with repect to systematic science (in biology, high throughput gene genome analysis, gene arrays, proteomics, etc.) and its associated automatization - "it is my belief that the authoring of papers will have to change. Practical science is moving towards Contract Research in a big way.... That scenery may create a new hierarchy of authors: conceptual vs. technical. The interpretation of data will be more central..." This may be explored in later article as it brings up many other issues.

One could add further credit distinctions to resolve this and other issues but if the credits are too complex, the solution will not appeal to anyone, be they journal editors, bibliographical database coordinators or readers. At least a two-line credit is simple and easy to understand and goes some way to curing the problem.


Copyright © BioinfoTools 2001, 2002—.   Last revised 12-Dec-02 12:27 AM (v3b).
If you have any problems with, or comments about, this website please read the Webmaster's page.
If you continue to have problems, email: webmaster@bioinfotools.com
This web site has been made & verified using BBEdit 6.5.2. Developers may find some tips & discussion on the Webmaster's page.