Writing a manuscript

1
Writing a manuscript

• Career Development Workshop Seminar Series
• Gary Westbrook
• Vollum Institute
• February 11, 2008

2
Writing a manuscript to maximize visibility for
your work
3

• What makes a good paper?
• Quality of the question
• Quality of the experimental design and execution
• Does it tell a story?
• Is it presented in a way that is accessible to
  the general reader?
• Concise, clear writing
• Learn basic grammar and English usage, but a
  style book, take a course
• The work should fit the Journal e.g. if you
  send every manuscript to Nature, your colleagues
  and the editors may wonder about your
  objectivity.

4

• Abstract
• It should give the reason why the work was done
  and the main findings. Remember most people never
  read your paper, if you are lucky they will read
  the abstract.
• Sentences such as The significance for
  Alzheimers disease will be discussed are
  worthless. If there is no conclusion, maybe you
  need to do more experiments.
• Depending on the journal, the reader should be
  able to understand the type of experiment that
  was performed from the abstract
  (electrophysiology, protein biochemistry,
  behavioral analysis, etc.)
• Highly speculative conclusions are risky. Your
  credibility is as important as getting the work
  published.

5

• Introduction
• The purpose of an introduction is to give the
  specific rationale and background for these
  experiments.
• An extensive literature review is not
  appropriate.
• The scope of the introduction should fit the
  content of the work. For example, an extensive
  discussion of the cost of Alzheimers disease to
  society is not appropriate for a paper on
  b-amyloid toxicity in cell cultures.
• Write only as much as needed. Some papers will
  need only very brief introduction, while others
  may require more explanatory material.
• Avoid tired phrases such as Glutamate is the
  major excitatory transmitter in the brain.
  Start with an idea that is specifically
  applicable to your work.
• Do not reiterate the same points (or phrases)
  that were made in the abstract or will be
  discussed in the discussion.

6

• Results and Figures
• Put the overall results in the text. The
  description of the specific example in the figure
  should be in the figure legend.
• Use only as many figures as needed.
• Size the figures appropriately with readable
  legends.
• Design figures so that they can be understood
  almost without looking at the text or legend if
  the reader is knowledgeable in the field.
• Avoid using the same figure layout for multiple
  figures. For example, 12 figures with multiple
  histograms is visually boring.

7

• Supplementary figures
• Data essential to the paper should go in the
  manuscript, not the supplement
• If you have more supplementary figures than
  figures, you may be sending the paper to a
  magazine instead of a journal!
• Legitimate supplementary figures should be
  supplemental, I.e. primarily of interest to a
  specialist or an illustration of a control that
  could be stated in a sentence in the text.
• As a reader, remember that supplementary
  information is rarely reviewed with the same
  rigor as the manuscript itsel
• Avoid using the same figure layout for multiple
  figures. For example, 12 figures with multiple
  histograms is visually boring.

8

• Discussion
• AVOID reiteration of the results
• Do not walk through discussion points in the same
  order as the figures were presented
• Pick several topics (e.g. 3) that you think are
  specifically addressed by the work and discuss
  those, e.g.
• Comparison to prior results
• Discussion of finding A
• Discussion of finding B
• Significance at another level e.g. to disease,
  to system or
• network function, etc.
• Keep the speculation to a minimum and save for
  the later parts of the discussion

9

• Dealing with Editors
• Be courteous and straightforward, but dont be
  shy about making your point.
• Follow the Instructions for Authors
• Dont try to oversell the message of the paper in
  the cover letter
• Assume the editor is an intelligent scientist
  (until proven otherwise)
• Dont distort the comments of the reviewers in a
  rebuttal letter. It will backfire when the
  editor looks at the revised manuscript and the
  reviewers comments.
• Think about what you have to say before you call
  or write an editor. Writing it down and editing a
  letter is much more effective, than a hasty
  e-mail dashed off minutes after receiving the
  reviews.
• Insults and letters filled with CAPITAL LETTERS,
  bold letters and EXCLAMATION POINTS!!! are not
  likely to get you where you want to go.

10
(No Transcript)
11
Picking a journal

• Assess the scope of your work, the potential
  audience, and the style of each journal
• Presubmission inquiries? - OK if you really have
  a question, but not a good strategy to grease the
  path
• A high average impact factor for a journal does
  not make your paper a good paper, or a highly
  cited paper (see skew distribution below)
• If you really care about citations for your
  paper, theres always GoogleScholar

From Colquhoun, Nature, 2003