The Human Genome Project took a battery of scientists thirteen years and three billion dollars to map a single genome. Twenty one years later, anyone with $150 can send away for a DNA analysis and get the results in a couple of weeks.
As gene sequencing technology grows increasingly rapid and more sophisticated, pharmacogenetics has advanced apace. Pharmacogenetics—an array method of probing specific positions in a patient's genome and computing which form of a specific enzyme they have—alerts prescribing physicians to drug incompatibilities. Meanwhile, Next Generation DNA Sequencing (NGS) uses tumor samples or blood to identify changes in patient DNA that alter specific genes, changing that region into cancer.
Larry Helseth, who has taught in the University of Chicago’s Master of Science in Biomedical Informatics, has developed software that takes full advantage of both swiftly evolving fields. As Translational Bioinformatician at the NorthShore University HealthSystem, Helseth works with a team of physicians, pathologists, pharmacists, programmers, and lab techs to improve patient care with bioinformatics.
For the past several years, he and his colleagues have been hard at work developing a new software they call Flype, which they presented in a paper for the American Journal of Medical Genetics. Flype analyzes the DNA sequence data coming off the DNA sequencers and helps interpret any changes, determines whether or not said changes are deleterious or have been seen before in other cancers, and identifies any drug that can be used to treat the patient based upon the changes in their DNA.
We started Flype because our pathology department was doing DNA sequencing of cancer samples, and the information coming back from the sequencer would indicate a variant in the position of a genome, but it didn’t tell us what that meant.Larry Helseth, Former MScBMI Instructor
“I'm not a pathologist or oncologist—I'm a biochemist who learned how to write computer code,” Helseth says. “We started Flype because our pathology department was doing DNA sequencing of cancer samples, and the information coming back from the sequencer would indicate a variant in the position of a genome, but it didn’t tell us what that meant. [To find cancer-causing mutations], we had to take the results from the sequence and run them through internal interpretation software and sort out all the things that didn’t change the protein, then bring in information about each of those positions.”
To simplify the process, the NorthShore team put all this information into a database, built a variant repository, and designed the software to interpret the results. Flype can optimize cancer treatment by recommending certain medications and warning against others. The software also handles pharmacogenomic tests that probe the results from a targeted microarray analysis test, aggregates the information acquired, then matches the results to translation tables developed by pharmacists and geneticists to warn physicians of drugs to avoid for their patient. For physicians, the process works something like this: a pathologist uploads the raw data from an array machine, the program generates results, NorthShore’s bioinformaticians prepare a summary of these results, a pharmacist signs off on them, then the program generates a report for the ordering physician and the patient.
“Flype presents the results of a cancer assay back to the pathologist on a webpage that takes 100 potential variants and summarizes them down to five or ten actionable ones,” Helseth says. “Our goal in pharmacogenomics is to put the information we get from looking at these genes into a patient’s medical records so that if a patient is prescribed the wrong drug, an alert will come through our system telling the physician we have genetic evidence that recommends they should not prescribe the drug and to choose an alternative.”
Flype, Helseth says, has similar applications in psychiatry. “We have a pharmacogenomics program at NorthShore, and many patients come through looking for a diagnostic odyssey trying to find the right antidepressant,” he says. Finding the right antidepressant through traditional methods “might take a couple of years—get an appointment, come back, and try another one. But by looking at the genes, we can often say what drugs won't work because certain enzymes are inactive.”
Due to its wide-ranging applications, a growing number of prescribing physicians are using Flype—just one of the new and emerging bioinformatic tools creating solutions across the biological sciences and medical fields. From expert instructors with deep backgrounds in health, research, and technology, students in the Master of Science in Biomedical Informatics learn to use the latest techniques for improving patient care.