NIH Launches Largest Human Genome Database

The National Institutes of Health has made a significant announcement: it has created the world’s largest database of human genomes. This repository aims to advance personalized medicine, which is an exciting prospect.

Scientists are already crafting clinical tools using genomic data from the All of Us program. One noteworthy development is the use of polygenic risk scores. These scores can estimate the chances of someone developing severe health issues, such as heart disease or breast cancer, starting in infancy.

However, there’s a major issue at hand. The current tools predominantly utilize data from individuals of European descent, making them less effective for other populations. In fact, for certain conditions, the predictions for people of color can be nearly random, almost like flipping a coin.

Researchers are actively working to address these biases by improving models and reaching out to underrepresented groups. There are concerns that without tackling these issues, health care inequality could worsen, hindering the potential benefits of this technology for chronic disease management.

Eimear Kenny, who directs the Institute for Genomic Health at Mount Sinai, emphasized the need for urgent attention to this issue. “It really warrants attention,” she said.

Polygenic risk scores are calculated by compiling numerous small genetic variations that could incrementally increase disease risk. By aggregating these variations, researchers can form a comprehensive score indicating a person’s susceptibility to prevalent diseases.

For individuals at the very high-risk end, these scores can offer crucial medical guidance, providing physicians with a significant head start to intervene. For instance, patients identified as having a high risk for cardiovascular issues might be advised to take cholesterol-lowering medications sooner to potentially avert heart attacks.

Yet, the effectiveness of these tools is heavily reliant on the quality of the data they’re based on. The UK Biobank, recognized as a key genomic resource, is predominantly homogenous, with about 94% of its participants being white. Similarly, the Million Veteran Program is made up of almost three-quarters white individuals.

Even the All of Us program, created to enhance data diversity, still consists of around 50% European participants, according to Alicia Martin, a geneticist at the Broad Institute. Moreover, the funding for this program is uncertain, with one major source potentially expiring soon and the budget already slashed significantly.

“You can’t just snap your fingers and overnight have a biobank from Africa, a biobank from India, a biobank from China that are all as large and open and comprehensive as the UK Biobank,” Dr. Martin noted. She added that while advanced statistical methods are useful, they cannot compensate for a lack of diverse representation.

Other health tools might suffer from the same ancestry bias. Integrated AI models based on genomic and biomedical information could yield skewed results, and geneticists searching for rare mutations may encounter genetic variations in patients of color that remain under-researched.

Blind spots in genetic research can hinder discoveries beneficial to wider populations. Take the PCSK9 gene, for instance. Researchers identified a rare mutation predominantly found in a small percentage of Black individuals that significantly reduces heart disease risk. This finding led to drug developments benefitting a broad range of ethnicities, but that crucial variant was only discovered because one study ensured a diverse participant pool.

The genetic diversity in African populations reflects a history of substantial genetic variation, which starkly contrasts with the reduced diversity seen in contemporary Europeans, who trace their ancestry back to a small African group that migrated many thousands of years ago.

If populations were to be chosen for genomic studies, some experts argue that Europeans would be the least representative, as stated by epidemiologist Jay Kaufman.

The most straightforward fix seems to be increasing the recruitment of people of color, a priority for the All of Us program. This initiative partners with local organizations, such as churches and health centers, to discuss and enroll participants from diverse backgrounds, even utilizing mobile clinics to reach rural areas. The program also shares health insights and genetic risks with participants, fostering a sense of reciprocity.

Yet, building trust is challenging, particularly given the U.S.’s historical injustices against people of color, such as the notorious Tuskegee syphilis study and unethical practices in research, which have understandably cultivated mistrust.

“Why would you contribute if you’ve been wronged in that way in the past?” Dr. Martin questioned, highlighting the need to address this mistrust before expecting full participation in genomic studies.

Meanwhile, researchers are adapting polygenic risk scores to use data more effectively. Modern modeling tools can now consolidate information from diverse sources, rather than relying solely on large European datasets. Institutions like Mount Sinai and U.C.L.A. have developed more inclusive biobanks with tens of thousands of genomes.

Efforts also extend to rapidly expanding biobanks in Asia, enhancing the accuracy of risk scores for East Asian individuals. Additionally, initiatives in countries like Peru and Qatar aim to gather genetic data from local populations, while the H3Africa program focuses on genetic research in African groups.

Moreover, academic networks have been tapping into non-biobank sources to broaden their data pools, utilizing projects designed to specifically include minority participants, like the ARIC study.

Despite these efforts, opinions vary regarding their sufficiency. Some experts stress that even an ethnically varied dataset may still fall short due to other influencing factors like socioeconomic status and access to healthcare. Others claim the significance of the bias might have been overstated in the first place.

Dr. Kenny, who has been involved in testing existing polygenic risk scores across diverse populations, expressed skepticism about quick fixes. “I don’t think these things are perfectly portable yet — and maybe never will be perfectly portable,” she remarked. “But there are things that are starting to narrow that gap.”