The fallacy of the national average in structured populations: clinical and population genomics across 9,768 Indian genomes

The human body is a vast library of instructions, written in a code of four chemical letters that spell out how we grow, how we fight disease, and how our bodies process the food we eat. For decades, scientists have been trying to read this library to understand why some people get sick while others stay healthy, and why certain medicines work for one person but harm another. To do this, they have built massive reference catalogs, essentially giant dictionaries of the most common genetic variations found in people around the world. However, these catalogs have a blind spot. They often treat large, diverse regions as if everyone there is the same, averaging out the details to create a single "national" picture. This approach works well for populations that have mixed freely for thousands of years, but it fails in places where people have lived in distinct, isolated communities for generations, keeping their genetic lines separate.

India is the perfect example of such a place. It is home to more than 1.4 billion people, but it is not a single, blended group. Instead, it is a mosaic of over 4,000 distinct communities, many of which have practiced strict endogamy—marrying only within their own group—for thousands of years. This social structure has preserved unique genetic signatures in each community, creating a landscape of diversity that is far more complex than a simple average can capture. When scientists look at the genetic data from India as a whole, they miss the specific stories hidden within these smaller groups. They might see a genetic variant as extremely rare because it is rare in the total population, not realizing that in one specific village or tribe, that same variant is actually quite common. This misunderstanding can lead to serious errors in medicine, causing doctors to misdiagnose patients or miss critical health risks entirely.

A team of researchers set out to fix this problem by taking a fresh, detailed look at the genetic data from the GenomeIndia Project, which recently sequenced the complete DNA of nearly 10,000 people from 83 different communities across the country. Instead of treating these 9,768 individuals as one big, mixed group, the researchers separated them back into their original 83 distinct communities. They wanted to see what happened when they stopped averaging the data and started looking at the fine details. What they found was that the "national average" was not just a simplification; it was a distortion that was actively hiding important medical truths.

The most striking discovery was just how many rare genetic variants exist in India. When the researchers looked at the data without separating the groups, they saw a certain number of rare changes. But when they peeled back the layers to look at each community individually, they found that the vast majority of genetic variations—more than 97 percent—were extremely rare, appearing in less than one percent of people. This is a much higher proportion of rare variants than what is seen in international databases that include South Asian people. The reason is that these rare variants are not spread evenly; they are concentrated in specific communities. A variant that might appear in only one or two people in a national pool could be found in dozens of people within a single isolated tribe. By averaging them out, the national view made these variants look invisible, when in reality, they are the defining features of local populations.

This invisibility has dangerous consequences for how genetic data is checked for quality. In standard genetic testing, scientists use a mathematical rule to spot errors. If a genetic variant appears in a pattern that doesn't match what is expected for a random, mixed population, the computer often flags it as a mistake and throws it away. The researchers showed that because Indian communities are not mixed but are instead distinct and separate, many real, medically important variants look like errors to these computers. For example, a gene variant that causes sickle cell anemia is a known, serious condition. In the national data, this variant appeared to break the rules of the quality check so badly that the computer would have automatically deleted it, thinking it was a glitch. By analyzing the communities separately, the researchers proved that these variants were real and critical, and that the standard way of checking data was actually deleting the very information doctors need to save lives.

The study also revealed a new way to decide which genetic changes are harmless and which are dangerous. Currently, doctors often rely on global databases to judge if a variant is a problem. If a variant is rare in the global database, it is often labeled as dangerous. However, the researchers found that many variants labeled as "pathogenic" (disease-causing) in global databases are actually common and harmless in specific Indian communities. They identified over 100 such variants that should be reclassified as benign because they are simply normal parts of the genetic makeup for large groups of people. Conversely, they found 55 other variants that are rare globally but very common in specific, isolated tribes. These are likely the real causes of disease in those communities, but they were being ignored because global databases didn't know they existed. The researchers proposed a two-track system: one track to clear away the false alarms of harmless variants, and another to protect and prioritize the real risks found in isolated groups.

Perhaps the most exciting finding was the discovery of "natural knockouts." In genetics, a knockout is a person who has two broken copies of a gene, meaning their body produces no working version of the protein that gene is supposed to make. Usually, finding such people is incredibly difficult because broken genes are so rare that you need to scan hundreds of thousands of people to find just one. But because of the long history of marriage within small communities in India, these broken genes have become more common in specific groups. The researchers found 123 different genes where they could identify people who were completely missing the function of that gene. Some of these genes are essential for life, but the people carrying these broken copies are healthy, suggesting that the body can sometimes survive without them. Others are linked to how the body processes drugs or fights disease. Because these people were hidden in the national average, they were mathematically invisible to previous studies. Now that they have been found, they offer a unique opportunity to understand how the human body works and to develop new treatments.

The research also mapped out how different parts of India have adapted to their specific environments over thousands of years. In the high mountains of the Himalayas, people have developed a genetic advantage that helps them breathe in thin air, a trait they inherited from ancient human relatives who lived there long ago. In the hot, humid forests of the south, other communities have genetic changes that help them resist malaria. In the north, where dairy farming is common, people have a higher frequency of genes that allow them to digest milk as adults. These adaptations are not spread evenly across the country; they are sharp, local features that disappear if you look at the whole map. A person from the mountains might have a very different genetic response to a drug than a person from the plains, even if they are both Indian.

The study concludes that treating India as a single genetic unit is a mistake that leads to poor medical care. The researchers demonstrated that to provide fair and accurate medicine, doctors and scientists must look at the specific community a patient comes from. They showed that by separating the data into its 83 original communities, they could uncover hidden risks, correct false alarms, and find new biological secrets that were previously impossible to see. This approach is not just about India; it offers a blueprint for understanding any part of the world where people live in distinct, long-standing communities. It proves that the key to unlocking the secrets of human health lies not in averaging our differences away, but in understanding them in their full, detailed complexity.