On Jun. 9, 2020, an international gnomAD team of over 100 scientists released its first set of discoveries in a collection of seven papers published in Nature, Nature Communications, and Nature Medicine. The flagship paper cataloged the genetic variation in both the protein coding and non-coding regions of human DNA. Included were more than 125,000 exomes (which include only the parts that code for proteins) and 15,000 whole genomes, from populations in Europe, East and South Asia, Africa, and more.

The team explored how likely certain variants are to cause a loss of function in the proteins produced from the gene. Protein-coding genes were categorized based on their ability to tolerate genetic variations without being disrupted or inactivated by them. This analysis found more than 443,000 genetic variants that were likely to cause a loss of protein function.

The second paper explored why mutations identified as likely to cause a loss of function don’t always cause the problems that might be expected. The team found that such variants are within segments of DNA that are often spliced out of the final mRNA copies of the gene used to produce proteins.

A third paper detailed the analysis of more than 433,000 structural variants in the human genome. Structural variants are changes that span long stretches of DNA, of at least 50 nucleotides. Structural variants were less likely to appear in protein coding regions than in non-protein coding regions. The team estimated that only about 0.13% of people carry a structural variant with any clinical significance.

The fourth paper explored how loss of function variations could be used to identify new drug targets. The fifth paper provided an example of how gnomAD could be used to validate drug targets. It analyzed the effects of loss of function variants in a gene called LRRK2, which has been associated with Parkinson’s disease. The results suggest that therapies to inhibit the LRRK2 protein would be unlikely to cause severe side effects.

The sixth paper described the impacts of variants in the region that sits immediately before the protein coding region of genes, called the 5’ untranslated region. The researchers identified specific genes where variants in this region could lead to disease. One novel variant they uncovered was tied to neurofibromatosis. Finally, the last paper showed how gnomAD could be used to analyze multi-nucleotide variants—clusters of two or more variants that are often inherited together.

The consortium’s next steps are to expand gnomAD to increase the number of genomes and diversity of populations included.