Twice in the past two days I’ve read sentences similar to this one implying that increased levels of the protein that shall be referred to as ‘X’ to avoid getting too specialized, is protective in models of Parkinson’s disease (PD):
“X mRNA levels are reduced in a number of neurodegenerative diseases …. since increased levels ameliorate behavioral defects and neuropathology of Huntington’s disease, Parkinson’s disease, and amyotrophic lateral sclerosis”.
The thing is that two papers published in 2012 showed that increased levels of X did not ameliorate the neuropathology of PD and actually caused some PD-like symptoms in model animals, like depletion of dopamine (that’s a big one). I was the first author of one of these papers and Carine Ciron from the EPFL was the other (1). Neither of these papers were cited in either article that referred to increased ‘X’ being protective against PD.
The issue is possibly clouded by one paper demonstrating that normalization of X levels rescued striatal pathology in a knockout model (2) and another showing that transgenic mice overexpressing X (without characterization of the level of expression) displayed increased resistance to the PD-inducing toxin MPTP (3).
In another example, I also recently read a review article where the authors mentioned all of the supporting evidence for increased levels of another protein (Y?) after exercise in humans and cited none of the studies demonstrating that this protein wasn’t increased after exercise.
This process of selective citation then becomes self-reinforcing as readers of those articles assume that the authors have done a good job in citing the literature and then often go on to cite the same papers without further thought or research.
There are a number of possible reasons why certain citations are left out. Journals have space limitations and that in many cases it is not possible to cite all of the relevant literature, however at the moment there are there are very few papers in either the X or Y field.
Also, people can be biased towards the literature that supports their theories (and the current trend for ‘hypothesis-driven’ rather than ‘exploratory’ research may have a hand in this) but as scientists we have a duty to be objective, present all aspects, and to not get too attached to our pet hypotheses.
Finally, sometimes people just lose track of the current literature, or a paper flies under the radar due to inappropriate cataloging or lack of promotion by the authors. However, with services such as Google Scholar, Pubcrawler and Citeulike that will deliver the relevant citations to your inbox or a daily, weekly or monthly basis, not citing the most relevant literature can come across as lazy or biased.
848 total views, 1 views today