Life Technologies' Ion Torrent is sponsoring a project at Carnegie Mellon University to develop open-source software that will help clinicians interpret human genomes, the institutions said this week.
Robert Murphy, director of the Lane Center for Computational Biology in Carnegie Mellon's School of Computer Science, will lead the multidisciplinary CMU team. Scientists at Baylor College of Medicine and at Yale University will provide whole-genome sequencing and patient data for the project.
Also involved will be Jaime Carbonell, director of CMU's language technologies institute, and Tom Mitchell, director of CMU's machine learning department.
Ion Torrent is funding the project during its first year with an undisclosed amount, and additional funding is expected to come from federal grants and other sources.
The ultimate goal is to develop a "doctor in a box" software that would diagnose disease, identify disease susceptibility, and predict drug response based on a person's genome sequence. Murphy said that new machine-learning tools will aid in the development of the software, which will be trained to analyze sequence data produced by Ion Torrent.
"It's an enormous undertaking," said Murphy in a statement, "but we are creating a framework that will allow us to tackle this problem one piece at a time and to do so at a scale that makes sense when all of those pieces are put together."
In the first year, the team plans to identify genomic features associated with a yet-to-be-determined disease or a patient population.
Scientists led by Richard Gibbs at Baylor's Human Genome Sequencing Center and Shrikant Mane at Yale's Center for Genome Analysis will perform whole-genome sequencing of patients and provide longitudinal medical records for them. These records will include treatments and outcomes as well as clinical test results.
Carnegie Mellon researchers will use machine-learning programs to analyze the anonymized data to find relationships between genomic data and clinical outcomes, and will incorporate information on genes and proteins from the literature. The result will be models that can be used to predict disease susceptibility and treatment response.
To support the project, Ion Torrent CEO Jonathan Rothberg will sponsor a scientific conference at Carnegie Mellon this summer or fall, called "Analyzing the $1,000 Genome."
Carnegie Mellon is joining a number of other projects and companies developing human genome interpretation software, including Knome, Omicia, GenomeQuest, Personalis, Cypher Genomics, the University of Utah, the Medical College of Wisconsin, and Harvard Medical School's Partners HealthCare Center for Personalized Genetic Medicine.