Most genetic testing of cancer cells focuses on individual point mutations, single spots in the genome where a base pair has changed, gone missing or been added. But these mutations don’t account for all cancers. There are many other structural variations in which large chunks of the chromosome undergo some kind of dramatic change and result in malignancies. “Now we realize that not just individual genes are important,” says Mikhail Kolmogorov, a scientist at the National Cancer Institute, “but how everything is arranged is important as well.”
Kolmogorov is developing computational methods to find these structural variations in biological data ranging from genome sequences to the complete protein output of cells or organisms. His work may soon translate to cancer clinics, where an algorithm that Kolmogorov and his colleagues developed can detect structural variations in chromosomes that may drive as many as half of all cancers.
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
Jeffery DelViscio
Another algorithm, developed during Kolmogorov’s doctoral research, is considered the state of the art for accurately piecing together long reads of a genome from fragmented sequencing information, a challenge akin to solving a million-piece puzzle. Kolmogorov and his team have also developed an algorithm called Severus, which works efficiently to reconstruct the genetic jigsaw. In a test in 17 pediatric leukemia patients, Severus confirmed five genetic drivers already known from traditional testing, as well as four additional genetic drivers that were missed by standard methods.
The goal, Kolmogorov says, is to build personalized genomics testing that will become routine for every cancer patient. That will take time, he says, but the methods are already useful for rare cancers with no known genetic biomarkers. And as it happens, childhood cancers tend to fall into this category. Better genomics testing would enable oncologists to perform more accurate risk assessments and formulate targeted treatments, Kolmogorov says. And that means algorithms like Severus could be literal lifesavers for the youngest cancer patients.
This article is part of “The Young American Scientists,” an editorially independent project that was produced with financial support from Regeneron.
