Tens of thousands of “dark” genes may still be missing from our human genome record. A global consortium of researchers has confirmed that these elusive stretches of genetic material can encode small proteins, some of which play a role in disease processes such as cancer and immunology.

These may explain why past estimates of our genome size are much larger than what the Human Genome Project found 20 years ago.

A new international study, still awaiting peer review, shows that our library of human genes continues to work as advances in technology reveal more subtle genetic traits and ongoing research reveals gaps and errors in the record.

These forgotten genes are hidden in regions of our DNA that are not thought to code for proteins. These stretches were once considered “junk DNA,” but it turns out that small pieces of these sequences are still used as instructions for miniproteins.

Institute of Systems Biology proteomist Eric Deutsch and his colleagues searched the genetic data of 95,520 experiments to find parts of the protein-coding sequence, finding numerous proteins. These include studies using mass spectrometry to examine small proteins, as well as catalogs of protein fragments detected by our own immune system.

Instead of long, well-known codes that trigger the reading of DNA instructions to make a protein that points to a gene’s starting point, these “dark” genes are preceded by shorter versions that have eluded scientists.

Despite these missing pieces in the starting sequences, non-canonical open reading frame (ncORF) genes are still used as templates to make RNA, and some of them are then used to make small proteins containing only a few amino acids. Previous studies have shown that cancer cells contain hundreds of these small proteins.

“We believe that the identification of these newly confirmed ncORF proteins is extremely important,” the team wrote in their paper. “Their proteins may have direct biomedical significance, as reflected by the growing interest in targeting such enigmatic peptides in cancer immunotherapy, including cell therapy and therapeutic vaccines.”

Some of the genes encoding these mysterious peptides are transposons floating around our genomes, including sequences inserted into us by viruses. Other researchers call this abnormal. For example, some proteins known to exist from mass spectrometry data have only been found in cancer samples, so the genes associated with them may not naturally belong to our bodies.

“Therefore, it remains possible that some ncORF peptides reflect abnormal proteins thought to exist outside the context of the canonical proteome,” Deutsch and his team explain.

The researchers found that at least a quarter of the 7,264 identified sets of these non-canonical genes were capable of producing proteins. This meant the addition of at least 3,000 new peptide-encoding genes to the human genome, and the team suspects there are tens of thousands more genes missed by previous proteomic methods.

“It’s not every day that you open a line of research and say, ‘We could have a whole new class of targeted drugs for patients,'” neuro-oncologist John Prensner of the University of Michigan told Elizabeth Pennisi. Science.

The tools the team developed will help other researchers continue to uncover more of this dark genetic material. This study is awaiting peer review at: bioRxiv.