Michael Levitt's Tips For Writing Nih Grants

Michael Levitt's Tips For Writing NIH Grants

(Copyright Michael Levitt, Oct 1989. Update Oct 1992, Mar 1996, Jun 2001, May 2004)

KEY ADVICE

Think about your science in a broad scholarly way. Read what others are doing, study their methods and implement them for your self. Brainstorm widely and come up with that special idea. This may seem like an obvious step, so obvious that I have just added it to these notes, but it really is the most important.

GENERAL

The reviewer has a huge load and has very little time for each proposal. Make your work as clear and transparent as possible.
Does the work have obvious biomedical importance? Stress any connection to biology and medicine.
Ask and answer these questions before starting to write

Have fewer aims but deal with them in depth. Be focused on solving a problem.
Go after a scientific problem and bring in all the collaborators needed.
Be sure to check the literature very carefully. The 24 Study Section members know a great deal more than you think they do.
Use the template forms that are available on the Web. Write using a Mac or PC and use a word-processor for which there are templates (Microsoft Word).

FORMAT

Have a "cartoon" figure that describes the project in a simple picture. Remember you have 10 minutes before the panel that votes on you!
Use a clear, readable type of 12 characters/inch, 6 lines/inch (12 point).
Use available space well and eliminate any padding or cramming. A short, well-written, comprehensive proposal is very appreciated.
Use as few abbreviations as possible and have a list of any abbreviation you must use.
Have headings labeled A1, A2, .. B1, B2.2, .. D1.1, D2, ... for easy reference. Make many sub-headings that summarize the section. Use bold face.
Include clear, relevant figures and tables in text (not in Appendix) with self-explanatory legends. Box these figure so that they stand out.
Label the figures by the section (e.g.. Fig. C2.1) which refers to them for easy cross-referencing. Placing figures in the text is tricky with all versions of Microsoft Word but it is much easier if one sets the format of the inserted picture object to have Layout Style “Square”.
Refer to references by name and date (e.g. Smith, 1988), so that the reviewer can instantly tell whether you have the right references.
Give titles of all papers in bibliography.
Consider using two columns per page with right-justification as this is easier to read.

ABSTRACT

TABLE OF CONTENTS

Have a detailed table of contents.
List appendix material in detail in section 3 of p.3, i.e. give paper references with title if there is space.

BUDGET

Do not expect funds for personnel that are "to be named".
Justify every budget item, especially equipment and foreign travel. Use specific brand names and give as many details as you can
Avoid excessive supplies, computer maintenance, computer time, etc.
List all personnel involved, even those not to be funded. List their positions and duties.
It is very difficult to get equipment in any year but the first.
In a renewal, do not try to get more than 10% more than your last year of funding (excluding equipment).

BIOGRAPHY

Mention the names of your supervisors for Ph. D., postdoc, etc.
Give all papers for last three years plus relevant earlier publications. Try to show recent productivity.
Separate refereed papers from book chapters. Start the list with the sentence "(selected from 45 referenced and published)" to show your total output.
Include papers that are "in press" and "submitted".

OTHER SUPPORT

List items clearly with all questions answered very specifically.
Be clear about your total % effort being less than 100%.
Have effort on each grant between 15% and 30%.
Avoid overlap, or if there is overlap, carefully justify the need for a new proposal.

AIMS

SECTION B - BACKGROUND

Show a good appreciation of the literature. Try to survey the field uniformly. You never know who will be on the Study Section.

SECTION C - PRELIMINARY RESULTS & PROGRESS REPORT

Have fresh preliminary results before writing the application.
Publish during the previous funding period (at least 2 paper/year).
List papers published during previous grant period prominently; specify overlap with other grants, if any. Use boldface for the author or authors supported by the grant in the publication.
In a competitive renewal, be very clear about which papers mention which grant. Be very clear about who was supported by which grant.
Summarize the results of each paper. Be sure to add new results not yet published. Be very specific about what result is in what paper.
If this is a competitive renewal, tabulate staff with dates of service and % effort . Give the grant evaluation and priority score from the previous review. Indicate the extent of requested funding that was actually granted to indicate that you managed on less than needed.

SECTION D - EXPERIMENT DESIGN & METHODS

Be humble and self-critical.
Do not be overly ambitious.
Limit the number of projects. Be focused and realistic. If you list more than is possible then the conclusion must be that you do not intend to do anything properly.
Give evidence to support anything that is unusual and imaginative.
Do not make difficult things seem too easy. Be realistic and convincing.
Do not get into areas in which you are not expert.
Divide each subsection into: "specific objectives"; "rationale"; "research plan"; "expected significance".
List the strengths of the proposal.
List possible pitfalls and how you plan to deal with them. This is very important.
Give a specific time-table.

OTHER

Have strong letters from any collaborators. The letters should not just be an generic endorsement.
Check spelling of names given in references. Show you care about the proposal and you will be considered a careful scientist. Avoid careless errors as these will be seen as indicative of sloppy science.