Machine Learning Untangles the Epigenetics of Cancer
A new study uses machine learning to identify genes that are involved in cancer even without changes in their DNA sequence
A tale of rogue cells
Cancer is the second leading cause of death worldwide (on global average, about 1 in 5 of us will have to face these rebellious cells). When a cell begins to divide uncontrollably and foregoes the normal rules of growth and development, the result is cancer. (For a short story about the life of a cancer cell, check out ‘Divide and Conquer’.)
At its root, the cause of cancer is a collection of mutations. Copying DNA during cell division is an incredibly accurate process, but it’s not completely flawless. Sometimes the resulting errors, or mutations, can accumulate into a cancer-promoting change in the cell’s many processes. (Of course, not all mutations are bad, but that’s another story.)
External substances can also cause mutations. Excessive exposure to UV light (such as blaring sunlight or, even worse, tanning beds) is strongly linked to an increased chance of skin cancer, and things such as air pollution and processed meat, are implicated in lung cancer and colorectal cancer, respectively. Our gut microbiome influences cancer risk as well…
