Using bioinformatics tools, cancer researchers can now search for common protein markers shared among afflicted patients. The process involves tracking thousands of proteins over a course of the disease and identifying the ones that map to patient survival. In a recent study, researchers believe they have a better chance of finding these relationships with a program inspired by Google’s famous PageRank algorithm.
Last week, the Txchnologist covered the scientists’ unconventional method. Christof Winter, one of the study’s researchers and a computational biologist at Lund University in Sweden, explained how cells respond to protein and gene interactions:
“A cell integrates many different inputs from the inside and outside and makes decisions based on them — grow, divide, migrate, differentiate, and so on. These decisions are mostly the result of proteins talking to each other, and if we want to predict what the cell does next, we have to, besides measuring the protein levels, take into account and better understand these networks of interactions.”
The researchers attempted creating their own algorithm before realizing that Google’s PageRank algorithm solved essentially the same problem for the web. They modified the algorithm somewhat, and came up with NetRank, a code that analyzes the relationship between proteins and gene expression.
Initially they used it to study pancreatic cancer. They found that out of 20,000 proteins they looked at, seven seemed to correlate most strongly with the how aggressive the cancer became. That information could then be used as criteria for patient treatment.
Significantly, the researchers found that NetRank was able to produce a prognosis that was 6 to 9 percent more accurate than conventional medical practices. Unfortunately, the program only applies to patients already diagnosed with the disease and does not allow for early detection. And more testing is required before the software can be used in real-world clinical environments.
According to their paper, written up in the PloS Computation Biology journal, the scientists view the application as a tool for medical professionals to improve individualized care. The researchers conclude that the technology can be used in a clinical setting to help decide if a cancer patient should receive chemotherapy. “Reliable prediction of survival and response to therapy based on molecular markers bears a great potential to improve and personalize patient therapies in the future,” they write.
Beyond predicting patient outcomes, information gleaned from NetRank could assist in the development of new cancer fighting drugs. For example, the program identified a protein named STAT3, believed to shorten the survival rate of a patient. With the protein identified, pharmaceutical manufacturers can begin to develop and test STAT3-inhibiting drugs, which might slow or reverse the cancer’s progression.