Sunday, 7 October 2018

New DNA Tool Predicts Height, Shows Promise for Serious Illness Assessment

A new DNA tool developed by Michigan State University can accurately predict the size of people and, more importantly, assess their risk for serious diseases such as heart disease and cancer.

For the first time, the tool or algorithm provides predictors of human characteristics, such as height, bone density, and even the level of education that a person might achieve solely on the basis of their genome. But the applications can not stop there.

"While we validated this tool for these three results, we can now use this method to predict other complex features related to health risks such as heart disease, diabetes, and breast cancer," said Stephen Hsu, lead study investigator and vice president of research and graduate studies at the MSU. "That's just the beginning."

Other applications have the potential to dramatically advance the practice of precision health. Doctors can intervene as early as possible in patient care and prevent or delay diseases.

The study, presented in the October issue of Genetics, analyzed the total genetic makeup of nearly 500,000 adults in the UK using machine learning, with a computer learning from data.

In validation tests, the computer has predicted the height of each human being within about an inch. Although bone density and predictors of educational attainment were not accurate, they were accurate enough to identify upstream individuals at risk of having very low bone density associated with osteoporosis.

Traditional genetic tests typically look for a specific change in a person's genes or chromosomes that may indicate a higher risk of developing diseases such as breast cancer. The Hsu model takes into account many genomic differences and creates a predictor based on tens of thousands of variations.

Using data from UK Biobank, an international source of health information, Hsu and his team used the algorithm to assess each participant's DNA and teach the computer to eliminate these significant differences.

"The algorithm examines each person's genetic makeup and size," Hsu said. "The computer learns from each person and ultimately creates a predictor that can determine just how big they are from their genome."

Hsu's team will continue to improve the algorithms while using larger, more diverse data sets. This would further validate the techniques and continue to help demonstrate the genetic architecture of these important traits and disease risks.

With more computing power and lower DNA sequencing costs, what used to be five to ten years is now much closer to this kind of work, Hsu added.

"Our team believes that this is the future of medicine," he said. "For the patient, a genomic test can be as simple as a cheek swab, costing about $ 50. Once we've calculated the predictors of genetic disease, early intervention can save billions of treatment costs and, more importantly, save lives. "

No comments:

Post a Comment