Looking at genomes from a new angle reveals patterns hidden in seemingly random differences

This story is part of our series, "Unsung cellular heroes." To read more stories in the series, click here.

Researchers often compare genomes of different populations—such as people who have a disease versus people who do not—in order to learn about those populations’ genetics—such as what genes may cause or contribute to a disease. Some diseases, such as cystic fibrosis and Huntington’s disease, are caused by changes to a single gene, and so it’s easy for researchers to spot the consistent difference between the groups and identify the disease-causing gene. Other diseases are more complex, and sometimes there are no clear patterns in the genomes of the affected population to point researchers towards the relevant genes. However, just because the patterns aren’t readily apparent does not mean they are not there. Work from Whitehead Institute Member Olivia Corradin and collaborators shows the value of taking a second, deeper look at seemingly random differences between individuals’ genomes, in order to decipher the hidden patterns that point to disease-associated genes.

Corradin and collaborators used their approach to study opioid use disorder, or the chronic and detrimental use of opioid drugs. Opioid overdose is a leading cause of accidental death in the United States. If researchers could identify the genetic risk factors for developing opioid use disorder, that information could be used to guide pain medication prescription and provide support to people at risk in order to prevent addiction and overdoses. However, opioid use disorder is complex, meaning that it is caused by some combination of changes to many different genes plus environmental factors. Each person affected by opioid use disorder may have a different set of contributing factors, making it harder for researchers to identify consistent differences in the genomes of people who have the disorder versus those who do not. Because of this complexity, and because researchers only have access to a small number of genomes from people with opioid use disorder, thus far they have not found strong enough patterns to identify the most important contributing genes.

In order to find useful patterns of genomic differences, Corradin and colleagues decided to look at the regions of DNA that regulate genes, rather than looking at the genes themselves. These regulatory regions, which increase the activity of target genes, are themselves regulated by the addition of chemical tags. The researchers looked for regulatory regions in which people who had died of an opioid overdose had lower levels of one such chemical tag than the average amongst people with no history of opioid use disorder. Lower levels of the tag would lead to lower activity of the regulatory region’s target gene. The researchers performed this search in the genomes of 51 people who had died of overdoses, and found that each one had many regulatory regions in their genome with lower-than-average levels of the chemical tag. However, there was little overlap in which regions those were from person to person. So far, the researchers had not found a pattern.

However, Corradin and collaborators looked a level deeper. They asked what genes these regions regulated, and that’s when a pattern emerged: many of the regions with lower levels of tagging targeted the same genes. For example, say there are five regions in the genome that each regulate the activity of gene A. In many of the people who had died of overdoses, at least one of those five regions had a lower-than-average level of the chemical tag. This means that even though the specific region was different in each individual, the individuals’ genomes had something in common: a change in regulation that should lower the activity of gene A. This commonality suggests that gene A may be relevant to opioid use disorder.

With this approach, the researchers identified five genes as having strong associations with opioid use disorder. Further research can confirm whether these genes contribute to the disorder (or are affected by it after the fact). By using a new approach to analyze the same data, the researchers were able to find hidden patterns in the genomes of people with opioid use disorder and narrow in on these promising disease gene candidates. The result of looking at the same sequencing data from a different angle is that researchers may be able to better understand and perhaps reduce the incidence of opioid use disorder and associated overdose deaths.