Growing up in a small town in Guangdong Province, Xueqiu Lin never met any scientists. But when she went away to high school and saw for the first time the moon and the stars through a telescope, she knew she wanted to be one.
Lin didn’t end up studying deep space. But her work does involve probing the depths of biological dark matter to unravel its mysteries.
Only 1%-2% of the human genome codes for proteins. The rest, scientists are coming to realize, cranks out loads of RNA molecules that don’t stray far from the nucleus and regulate gene expression. Researchers have slowly been able to piece together lists of regulatory elements that act on different genes, but not a systematic way to study how different elements linked to the same gene cooperate or counteract each other.
At Stanford, Lin developed a machine learning model to decipher a general rule for enhancers — elements that boost gene expression — using reams of data generated by high throughput CRISPR screens. Now an assistant professor of computational biology at the Fred Hutchinson Cancer Center, she plans to extend this research to other elements of the non-coding genome.
Such work has important implications for improving the precision part of precision medicine. More than 20% of cancer-linked genes are regulated by multiple enhancers. Knowing how they work together will allow for the development of more accurate cancer risk prediction tests.
“The key is understanding how they interact with each other to form a regulatory network,” said Lin. “The whole is more than the sum of its parts.”
— Megan Molteni
