We probably all agree that a publication bias against negative studies will severely distorts our opinion. To repeat an earlier Nature letter
Why negatives should be viewed as positives … This filtering of results undoubtedly biases the information available to scientists (see, for example “Null and void” Nature 422, 554–555; 2003). And communication is at the heart of science.
Here is an email that I received from the editor of Immunogenetics Ronald Bontrop (pp D. Devine – sender did not decline publication here)
I’m sorry to disappoint you a second time, but I stand by my original decision.
The three main reasons for this are:
1. None of the references cite Immunogenetics;
2. The journal no longer publishes negative associations;
3. The manuscript does not comply with the criteria set out by Manly (2005) Reliability of statistical associations between genes and disease. Immunogenetics 57: 549-558.
Whatever you think about argument 1. or 3., argument 2. should lead to a boycott of this journal.