STIX SV Index
For years, researchers have tried to quickly and accurately identify the portion of DNA that causes genetic diseases such as cancer
"Understanding cancer requires identifying genetic changes that force a patient's healthy cells to grow uncontrollably," said Assistant Professor Ryan Layer
The work is part of ongoing research in Lescher's lab in the Department of Computer Science and Frontiers in Biology, which uses algorithms to decipher very large genomic datasets
The process, described in a new paper in the journal Nature Methods, aims to quickly determine whether a particular gene sequence is common or rare, and whether it is likely to cause cancer in those particular tumor cells, among other things.
"A mutation hidden somewhere in the genome of a cancer patient's tumor encodes instructions for how the tumor begins to grow uncontrollably," said Layer
By counting how often a sequence occurs in healthy people to help determine whether it's a disease-causing mutation, the paper's lead author, Murad Chowdhury, a scientist in Ryder's lab, said: Not a new idea
The main challenge for the team was computational, Chowdhury said — reorganizing the massive dataset so that it only took a second to search for the information it needed
"Our technology simultaneously reduces data storage requirements and increases query speed, so analysis that would otherwise take months can be performed much faster," Chowdhury said
Researchers do a lot of work every year to analyze and categorize tumor information, Layer said
"This tool is designed to improve access to data and give users quick access to useful and accurate answers," he said