Interesting story about Nicholas Tatonetti & his New Computational Tools to Study Drug Effects
Courtesy Genomeweb & Julia Karow
Studying for a dual degree in mathematics and molecular biology, Nick Tatonetti became interested in using computational models to study biology and make sense of its massive datasets. As a bioinformatics PhD student at Stanford, he developed new statistical models and computational approaches for analyzing drug effects and drug-drug interactions.
At Columbia, Tatonetti is now focusing on molecular mechanisms of drugs. "We can actually think of each time a patient is being given a drug as an experiment," he says. "When the drug goes into the human system, it interacts molecularly, and then phenotypes come out of this system," which can be connected to molecular mechanisms in new ways.
In particular, he is developing techniques that use clinical data to develop networks that highlight interactions between different systems in the human body, such as two organs. "And once we know that certain gene products or pathways are working together at a systems level, we can start integrating all of the molecular data we know to fill out a complete picture" and to predict phenotypes from, for example, RNA expression data or genotypes, he adds.
Paper of note
A study revealing a new drug-drug interaction by using a machine learning algorithm Tatonetti designed that scanned through millions of adverse event reports from the Food and Drug Administration appeared in Science Translational Medicine this year. That is his favorite paper, he says, "because it represents the most comprehensive innovations on both the algorithmic side and the evaluation side."
Looking ahead
Tatonetti says that over the next five years, there will be major advances in using genetic data in medicine. "It may not be that we need to store lots of genetic data in the clinical record, but that we need to understand the genetic data better so that we can provide results and suggestions to physicians," he says. Personalized medicine will not only involve genotypes, though, but also environmental exposures. Putting genetics into the context of these, he says, will increase its clinical impact.
And the Noble goes to…
If he were to win the Nobel, it would be for developing a new scientific approach, based on informatics methods and computational methodologies, that discovers a fundamental biological or medical truth. "Where that truth may be, in what system, in what disease, I could not say, but I love the idea that we're in the midst of changing the way that we do science," Tatonetti says. "That's what informatics is about, advancing the tools of the scientific method itself."
Studying for a dual degree in mathematics and molecular biology, Nick Tatonetti became interested in using computational models to study biology and make sense of its massive datasets. As a bioinformatics PhD student at Stanford, he developed new statistical models and computational approaches for analyzing drug effects and drug-drug interactions.
At Columbia, Tatonetti is now focusing on molecular mechanisms of drugs. "We can actually think of each time a patient is being given a drug as an experiment," he says. "When the drug goes into the human system, it interacts molecularly, and then phenotypes come out of this system," which can be connected to molecular mechanisms in new ways.
In particular, he is developing techniques that use clinical data to develop networks that highlight interactions between different systems in the human body, such as two organs. "And once we know that certain gene products or pathways are working together at a systems level, we can start integrating all of the molecular data we know to fill out a complete picture" and to predict phenotypes from, for example, RNA expression data or genotypes, he adds.
Paper of note
A study revealing a new drug-drug interaction by using a machine learning algorithm Tatonetti designed that scanned through millions of adverse event reports from the Food and Drug Administration appeared in Science Translational Medicine this year. That is his favorite paper, he says, "because it represents the most comprehensive innovations on both the algorithmic side and the evaluation side."
Looking ahead
Tatonetti says that over the next five years, there will be major advances in using genetic data in medicine. "It may not be that we need to store lots of genetic data in the clinical record, but that we need to understand the genetic data better so that we can provide results and suggestions to physicians," he says. Personalized medicine will not only involve genotypes, though, but also environmental exposures. Putting genetics into the context of these, he says, will increase its clinical impact.
And the Noble goes to…
If he were to win the Nobel, it would be for developing a new scientific approach, based on informatics methods and computational methodologies, that discovers a fundamental biological or medical truth. "Where that truth may be, in what system, in what disease, I could not say, but I love the idea that we're in the midst of changing the way that we do science," Tatonetti says. "That's what informatics is about, advancing the tools of the scientific method itself."
Comments