BBC Article Title: Do Friends Have Similar Genomes?
Secondary piece URL:
http://www.bbc.com/news/science-environment-28295969
Primary piece URL:
http://www.pnas.org/content/early/2014/07/10/1400825111.full.pdf+html
Summary of BBC Article:
This article describes the findings of a study that claims on average, friends share 0.1% more DNA than they do with strangers. The study analyzed nearly 500,000 single nucleotide polymorphisms in the genomes of the participants of the Framingham Heart Study. The article includes quotes from the authors who explain their conclusions and also statements from scientists who are skeptical of the implications for the data that the two authors of the study, Christakis and Fowler, collected. With data collected from a small community in the United States, the authors calculated a kinship coefficient, which they found to be slightly higher amongst friends than strangers. Scientists such as Bowden and Charney suggest that there are other factors that affect these results, like types of population stratification, which could be confounding. Charney says that it is not a safe assumption that everyone in the study was not related, which if they were, the results would not be meaningful. However, Fowler claims that they excluded all people that were related. Charney also points out that the method of the study may not be able to account for factors that lead to friendship that would also lead to a correlation in genotype.
Primary Source Summary:
The article claims that ‘”friends’ genotypes at the single nucleotide polymorphism level tend to be positively correlated,” or homophilic, and that certain genotypes among friends tend to be negatively correlated, such as the immune system gene set. They then suggest that, because hemophilic genotypes equate to positive selection, hemophilic genotypes may behold a fitness advantage in human evolution, yet it is unknown what role genotypic correlation plays.The authors propose four reasons why people may tend to interact with people who have more similar genes to them than they do with a stranger and a smaller number of reasons why friends may exhibit heterophily in their genotypes.The scientists analyzed 466,608 SNPs in 1,932 subjects who were in one or more of 1,367 friendship pairs. They claim to have used strict control for population stratification for their final results, as they did not at first. Density plots show that friends have significantly higher kinship coefficients than strangers and that friends also have lower proportions of opposite genotypes. In a Genome-wide association study with strict control for population stratification, friends had more homophily and heterophily than strangers. In the conclusion of the article, the authors suggest that human evolution may be accelerating because the genes of people surrounding an individual may affect the fitness advantages of the individual’s own genes and people frequently interact with non-relatives.
Reflection
The information in the BBC article technically did make false claims, but it left out a few point present in the primary article that I think were important to understanding the implications and methods of the study. So, BBC did an average job of reflecting on the scientific finding. For obvious reasons, the BBC article hardly contains any data regarding the study and mainly focuses on the perspective of skeptics of the authoring scientists and the study at hand. The BBC article did not mention anything about what the finding might suggest about human evolution, which was a central theme in the scientific paper. Also, the BBC article did not mention the negatively and positively correlated genotypes present among friends, which may have made the study seem more well-rounded and thorough if included. I would improve the BBC article by talking more about the methods and data that the authors collected and also that they aren’t making concrete far-fetched conclusions about their findings. I think that the aspect of human evolution, which was the main theme in the conclusions drawn from the data, was very underrepresented. Instead of criticizing the methods of data collection, which were clear-cut in the real article, the BBC article could have talked about how the finding that friends have more similar DNA than strangers may be an agent of human evolution. To me, this would have made the article a lot more interesting and relevant.
I like your suggestion that the popular article should have included more information on the methods and data. They could also have included which genes the scientists found correlation in. I agree that the BBC article only shows a portion of the picture of the actual study.