Russian Scientists Develop Algorithm to Identify Colorectal Cancer Vulnerabilities
Researchers in Russia have developed an algorithm that can identify genetic targets for treating colon and rectal cancer within hours. The method has already identified 12 promising DNA targets. The technology could accelerate the development of new drugs and help physicians select therapies more precisely.
Colorectal cancer is one of the most challenging malignancies. It is the second leading cause of cancer death worldwide. More than 1.9 million cases are diagnosed globally each year, and nearly half are fatal. Tumors can adapt to treatment and leave behind dormant cells that can later trigger recurrence.
Researchers at Central University in Moscow have developed an algorithm that produces a prioritized list of target genes within hours. It works like a navigator through the genome: instead of wandering through a maze, researchers and oncologists can immediately see potential directions for therapy.
How the Algorithm Works
The approach is based on identifying mutations in noncoding regions of DNA that regulate gene activity. One such regulator is the long noncoding RNA SNHG1. Its levels rise sharply as colon and rectal cancer develops. But researchers previously did not know exactly how this regulator affects the tumor or which genes it acts on. The RNA’s considerable length and thousands of potential targets made the task difficult.
The Russian researchers developed an algorithm that considers several factors simultaneously, including how DNA and RNA interact, markers of gene activity, and the locations of super-enhancers – specialized sequences that can amplify a gene’s activity many times over.
“Previously, biologists had to use trial and error to determine which genes a particular long noncoding RNA might activate. We proposed an approach that produces a testable hypothesis within a few hours of computation,” said German Ashniev, a researcher at Central University.
12 Target Genes
The algorithm identified 12 promising target genes whose activity was elevated in patients’ tumors compared with healthy tissue. Among them was TOP1, a gene already known as a major chemotherapy target. This finding confirms that the method proposed by the Russian researchers works effectively.
“This result shows that the approach is not merely theoretical. We found a real gene that is already targeted by drugs. That means the algorithm can be used to search for new targets,” the developers said.
Why Does the Cancer Return?
The central challenge of colorectal cancer is its ability to adapt to treatment. During chemotherapy, most tumor cells die. But the most resilient cells remain within the tumor – those that are resistant to therapy from the outset. They adapt and continue multiplying, producing a new tumor that is even more aggressive and resistant to chemotherapy.
For a complete cure, a drug needs to act selectively, targeting the genes that the remaining cells use to protect themselves. But the human genome contains thousands of genes, making a manual search for the right target slow and labor-intensive.
The algorithm developed by Central University researchers addresses this problem. After just a few hours of computation, it produces a short list of priority genes. This can help physicians select treatment more precisely for an individual patient.
What Comes Next?
The developers’ next step is to test the identified targets experimentally and determine how strongly they actually influence tumor-cell behavior. If the findings are confirmed, individual genes on the list could provide a basis for further drug research.
The algorithm itself can also be developed further. A similar approach could be applied to other types of cancer and biological processes in which researchers need to identify connections among large numbers of genes and molecular mechanisms.
For patients, the ultimate significance of this work is concrete: the more scientists understand about how a tumor behaves and how it resists treatment, the more opportunities they have to strike it precisely where it is most vulnerable.
