On his Zika Diaries blog, Dr. Vincent Racaniello has a new post: Zika Critique. He tells us a lot about how science gets done, or doesn't get done. Excerpt:
In the last column I wrote of our first unsuccessful attempt to secure funding for our work on Zika virus. Since then we have received the reviewers' comments and now we can determine what they did not like.
Applications for research support under the NIH R21 mechanism are reviewed by a study section consisting of ~20 experts in the field. Our application, entitled 'The neurovirulence of Zika virus isolates', was reviewed by three individuals who prepared written critiques. Each reviewer scores (from 1-9 with 1 being the best) the grant application in five different areas: significance, investigator qualifications, innovation, approach, and environment. After the entire panel votes on the scores, the final impact/priority score is calculated by averaging the numbers and multiplying by 10. Currently proposals that are scored over 30 are not funded; ours received an impact score of 41.
Two of the three individuals assigned scores of between 1 and 3 for each of the five different areas; they were enthusiastic about the proposal and had few criticisms. Unfortunately the third reviewer did not like the proposal and scored it from 2-6 in the five different categories.
The goal of our proposal was to study the neurovirulence of different Zika virus isolates in brain slice cultures from embryonic mice. We wish to understand how the virus affects brain development, and whether Zika virus isolates differ in their ability to disrupt the brain.
The main criticism from the negative reviewer was that it is not clear how results obtained in mice would apply to humans. If this criticism were valid, then no virologist would be using animal models to study viruses that cause human disease. They are called 'models' for a reason: the findings might or might not be predictive of what happens in humans. But animal models are invaluable because experimental infections of humans with viruses are not done except with a few viruses that cause little or no disease. Furthermore, experiments in animals can reveal mechanisms of pathogenesis which might be validated by observations of natural infections in humans.