A computer program is presented that uses artificial intelligence to analyze DNA and predict complex human traits, including height, propensity to osteoporosis, and expected educational attainment.
American scientists have proposed an algorithm that can interpret DNA analysis data and predict growth, bone density, and even educational level of a person. “While we have confirmed the instrument's workability for these three specific cases, we can now apply the same method to predict other complex characteristics associated with the risk of cardiovascular disease, diabetes and breast cancer,” says Stephen Hsu of the University. Michigan State, lead author of an article published in Genetics.
In their work, the scientists used anonymized personal data and DNA sequencing results for 500 thousand people in the UK, contained in the open database of the UK Biobank. A neural network was prepared, capable of analyzing single nucleotide substitutions, many minimal differences in the genomes of different people. With the necessary training, she was actually able to predict a person's height using DNA data to within a couple of centimeters.
The accuracy of predicting the level of education and bone density was not so high, but the extreme options - the risk of developing abnormally fragile bones or serious difficulties even with schooling - were predicted quite confidently by the neural network. However, the authors continue to improve their program and work on expanding the data array for training the neural network. This should improve outcomes and expand the use of a new instrument that may one day become an important part of the medicine of the future.
Sergey Vasiliev
