Racial Bias Found In Genetic Databases
A study has revealed a racial bias in the two leading genomic databases, used today by leading clinical geneticists, favoring European ancestry.
The study was published October 11th in the journal Nature Communications. The principal investigator was Dr. Timothy O’Connor, PhD from the University of Maryland School of Medicine. Other parts of the research team came from Johns Hopkins University, the University of Colorado, and the Henry Ford Health System.
The article about this new study stated that recognizing the bias in these two databases could lead to more cost-effective, targeted therapies. If the database is biased towards European ancestry, it may take a lot more work and thus money (about $1000) to find an analysis of an African American’s genome compared to a European American. The result would not be guaranteed accurate either. This makes pursuing this research further appealing in terms of cost.
What the researchers did was study 642 entire genomes from the Consortium on Asthma among African-ancestry Populations in the Americas (CAAPA). In this 18-month study, 642 people’s genomes from African origins were studied, either from America, Africa, or the Caribbean. When compared to current databases, preference for European variants was shown than non-European variants. The article stated that the researchers “created the largest, high-quality non-European genome data set ever assembled.”
Why it’s important to include varying ancestries in genomic databases is it will create more accurate diagnoses. If somebody walks in with 25 percent African heritage, they may have different risks of diseases than someone with different ancestry. This is why it’s important to unbias these databases. It will lead to better accuracy in addition to cost effectiveness.
The person with 25 percent African ancestry may have different number of variants than someone with different ancestry. Variants are basically mutations in the human genome, which can lead to genetic diseases. However, humans are 99.5 percent similar in their genetic makeup.
The research could help clinical geneticists add more diversity to their clinical databases, resulting in more accurate and cost-effective diagnoses.