For many years biologists have developed a huge library of cell structures and their respective data. Study of specific areas of a living cell involves a phased approach that defines how the individual parts work with others, and cell labeling. Institute of cell Sciences Allen launched the first predictive 3D model of living human cells — Allen Integrated Cell — and it can “completely change the game”, according to scientists.
“It’s a new way to look inside human cells,” says Rick Horwitz, Executive Director of the Institute of cell Sciences Allen. “It’s like watching a whole cage for the first time. In the future it will affect the discovery of drugs, the study of diseases and how we build fundamental studies involving human cells”.
To create a breakthrough model, the researchers conducted gene editing huge collection of living cells people to turn on fluorescent protein markers that highlight specific structures within cells. Scientists have made tens of thousands of images of fluorescent cells and used AI to create a probabilistic model that predicts the most likely shape and location of structures in every cell depending on the shape of the plasma membrane and nucleus.
The researchers then used these pictures to another machine learning algorithm that used what was learned from the cells with fluorescent labels, to search for cellular structures in the cells without fluorescent markers. Combining all data, the system could produce images that look almost identical to the images obtained using traditional fluorescence microscopy, which can be expensive and toxic.
“Until now our ability to see what is happening inside of human cells was very restricted,” says Michael Elowitz, Professor of biology, bioengineering and applied physics at Caltech. “Before, we could see only those proteins that are selectively celebrated. But Allen Integrated Cell — this is a perfect free lunch. We have a “buffet” of various proteins and organelles, nothing illuminating. This opens up entirely new and powerful way of studying cell biology”.
Earlier, scientists also learned to explore the work CRISPR outside of human cells in vitro.
AI proposed a new way to peer into the living human cell
Ilya Hel