Thanks to Skillshare for supporting this episode of SciShow. [ INTRO ].

There's a lot of buzz surrounding precision medicine, which is the idea that we can tailor the way we treat disease to every individual person's unique circumstances. And though it's a futuristic idea, it's already helping us to diagnose and treat some diseases, like certain types of cancer.

The dream is that someday, the information in our genomes — the sum total of our DNA — could transform healthcare as we know it for everyone. But there are limitations to overcome before precision medicine can reach its potential. One major hurdle is the lack of diversity in human genomic databases.

The vast majority of human genetic studies we've carried out since the 1990s were done on people of European ancestry. It's a massive oversight that's already leading to health disparities, and it's going to require some ambitious fixes. But first, how did this happen?

It's pretty bad science to base our understanding of the human genome on white-centric data, considering how many of the humans on this planet aren't white. But under-representation and exploitation of minorities in medical research is nothing new. In 1996, the National Institutes of Health introduced a policy to prioritize including more women and underrepresented minorities in clinical research.

But since then, less than 2% of over 10,000 cancer studies funded by the NIH have included enough minorities to meet their own guidelines. There's also the totally justified lack of trust from communities of color when it comes to research participation because of repeated ethics violations. The most infamous case is the Tuskegee Syphilis Study that began in nineteen thirty-two, in which researchers recruited hundreds of African-American subjects with syphilis, and then… watched, as the deadly disease progressed untreated.

Even /after/ they knew that penicillin could cure syphilis. President Clinton apologized for the decades-long ethics breach in nineteen ninety-seven, but that doesn't mean everything is fine now. For instance, a 2018 analysis revealed that African-Americans are disproportionately enrolled in clinical trials that don't require patients to give individual consent.

This exception to consent rules exists for people who are in too life-threatening of a situation to sign paperwork, but the disparity is still a problem. There's also just plain structural inequality throughout society that influences how medical research is carried out, l eading to undeniable healthcare gaps between white people and, well, everyone else. How does this apply to genetics research and precision medicine?

The Human Genome Project published the first human genome in 2003. At the time, sequencing a single genome was a landmark accomplishment. Since then, sequencing technologies have gotten better, faster, and way cheaper, leading to mountains of genetic data rivaling the amount of data generated by YouTube, Twitter, and even the entire field of astronomy.

And most of that data concerns populations of European ancestry. Which has led to a sort of snowball effect, as researchers tend to prefer to study the most well-characterized groups, and go on to do more studies in white people. One major problem is with the reference genome — used by researchers worldwide as, well, a reference.

A 2018 study on genetic data from over 900 individuals of African descent found that over 10 % of their DNA wasn't represented in the reference genome, most of which is derived from a small number of people. Now, it's /really/ important to note that none of this is to say that people of different races or ethnicities have dramatically different DNA. Other research has found that any given single person differs from the reference genome by well under one percent of their DNA.

It's just that different individuals have different differences -- and that adds up. The lack of diversity in genetic databases has already led to disparities in translating precision medicine research into diagnoses and clinical care. For example, a 2016 analysis found that African-Americans have been disproportionately misdiagnosed with a heart condition known as hypertrophic cardiomyopathy.

A gene variant that's more common in black populations had incorrectly been identified as dangerous. And the researchers suggested that including more African-Americans as controls in genetic studies could have prevented that mistake. Researchers have also raised concerns that basing tests for genetic diseases on our current body of knowledge leaves communities of color by the wayside.

For example, most of what we know about the genetics of asthma comes from studies of European populations. But studies have found those gene variants are found less often in many African-American and Puerto Rican individuals -- even though their incidence of asthma is higher than in white people. If you were to design a genetic test based on the presence or absence of those variants, you'd have something only white people can actually use.

Drug safety and dosage also come into play. Genetics can affect how well a drug works, the ideal dose, and the likelihood of adverse reactions for each individual. For example, warfarin is widely used as a blood thinner.

And variants in three genes can help doctors understand how patients will respond to it — and help them determine the right dose to avoid dangerous side effects. Except that's only the case in European populations. It doesn't explain how people of other ethnicities will metabolize the drug.

Which means they're at a greater risk of an adverse reaction. This lack of diversity isn't going to fix itself, but there are proposed solutions, and some are already in the works. The National Institutes of Health launched the All of Us research program in 2015 to collect genetic and health data from more than a million people living in the U.

S., with a focus on ensuring a diverse pool of subjects. Many non-European countries have already launched population-specific genome projects to produce reference genomes that are more relevant to those populations. Some experts have also suggested that scientific journals either require researchers to draw on diverse sample populations, or provide a darn good reason why not, as a requirement to publish.

Researchers will also need to reclassify genomic variants -- like the heart disease genes -- based on data from diverse populations. There's a newly revived push to implement culturally respectful strategies for recruiting underrepresented groups in research, with local ethics committees in the communities where research takes place playing a key role. These solutions have to happen within the larger context of tackling systemic racial disparities throughout society.

For instance, not only does the scientific community need more diverse data, it also needs to educate and retain more researchers of color, since diverse perspectives lead to more diversity in research design and participation. Tackling genomics' diversity problem is a massive undertaking, but it's going to need to happen if precision medicine is going to change healthcare for the better for all people. At its best, science embraces past mistakes to learn from them and do better in the future.

And that's also a mark of a good leader -- according to this leadership class you can take on Skillshare. In her course Leadership Today, Steph Korey teaches you how to use coaching and questions to grow your business. And that includes consistent feedback, asking questions of your team, and embracing your mistakes.

There are over 25,000 courses on Skillshare, from business to photography and design, so you're sure to find something for you in this community of more than 7 million creators. And an annual subscription is under ten bucks a month. Right now, the first 500 SciShow subscribers to use the link in the description will get a 2 month free trial.

So check it out, and see what you can learn! [ OUTRO ].