Saturday, March 17, 2018

Cell Biology Decodes Artificial Intelligence

Image: iStock

Artificial intelligence (AI) is an intricate series of neural networks, quickly reaching complexity beyond that of the human mind.
Operating as a black box, the process in which artificial intelligence systems deliver the output to questions such as “What animal is in this photo?” is greatly unknown. However, a study analyzing a neural network mapped with a basic yeast cell has given new perspectives into both cell biology and the inner workings behind artificial intelligence.
Neural networks are formed by layering “neurons” to complete all the individual tasks needed to compile an answer. Then, large databases are entered, responses given by the artificial intelligence system are verified, and the system is improved to yield more accurate results.
Ideker and his team have mapped 2,526 subsystems from the yeast cell into a neural network to create DCell. DCell is able to simulate cell growth and the impact of genetic mutations on yeast cells, introducing the potential for human cell simulation. Ideker explains how once this upgrade is achieved, the healthcare industry could be completely transformed, using cancer research as an example: “You could boot up the model with the patient’s genome and mutations, and it would tell you how quickly those cells will grow, and how aggressive that cancer is.”
Read the study at: https://www.nature.com/articles/nmeth.4627

No comments:
Write comments