AI Using Computer to Decipher Cells

We now is using computers to decipher data that might make sense in a microscopic scale.

Researchers have teamed up to create what is being called the first computer model that can predict growth of cells from noisy data. The model might be able to predict how a microbe behaves, such as how its population might change or shape and then evolve to solve a given problem, such as how malaria would affect a human.

In order to predict cell growth, the high-performance computing power was used. A computational model that takes into account transcription, expression and differentiation of human cells. To figure out the relationship between the three signals, the scientists employed an algorithm that translates these signals and predicts the growth of a cell.

The algorithm also described how different cells respond to environmental stimuli and categorized cells as reactive or non-reactive. Once the model is decoded, machines could then look at the patterns to get a better understanding of how cells grow.

Each cell is like a miniature chemical factory, creating a multitude of proteins in varying states. While each cell generates very tiny molecules, each molecule can evolve into a different structure depending on the environment. For example, a cell that is smaller than a bacterium is non-reactive, but a cell whose DNA is fragile will develop into a very close cell. In other words, in order to change a cell, scientists need a different environment.

We hope that the model will be used to predict whether or not a microbe will evolve to survive in an area of infection. For example, human could provide their colonies with a very cold environment, without causing any killing effect. But, if a microbe adapted to this environment, it could eventually eventually evolve into a healthiest microbe in the long term.

The algorithm will be more accurate than existing models for predicting cell growth by relying on just the enzyme signals, which are not always as accurate because they only focus on a single step. The algorithm also takes into account how a microbe's behavior affects its environment to get a more complete picture of cell growth.

Moreover, according to Vaziri, we will also be able to use this algorithm to see the relationship between different tissues and models using computer models.

This article refer to LiveScience