And that may give us more insight into how human diseases arise.

Most human diseases can be traced back to defective parts of a cell. For example, a tumor or tumor can arise due to incorrect cell division. And in a metabolic disease, harmful intermediates accumulate in cells. However, to better understand what exactly goes wrong in a cell, scientists first need to gain more insight into the different parts that make up cells. And so researchers, armed with artificial intelligence, set out in search of as-yet-unknown cell components.

Cell

Imagine a cell. You probably picture that colorful drawing from your cell biology textbook, complete with mitochondria, endoplasmic reticulum and the cell nucleus. But is that the whole story? “Absolutely not,” said researcher Trey Ideker. “Scientists have long realized that there is much more than we know about. But only now do we have a method to get a better look at a cell.”

Small smaller smallest

Researchers have been interested in mapping the internal workings of cells for years. But that is not without a struggle. Microscopes allow scientists to see down to the level of a single micron (a micron is one millionth of a meter). This is the size of some organelles, such as mitochondria. Smaller elements – think individual proteins and protein complexes – cannot be studied through a microscope. In that case, biochemical techniques are often called in, with which scientists are able to get down to the nanometer scale (a nanometer is one billionth of a meter, or 1,000 microns). But how do you bridge the gap from nanometers to microns? “That has been a major hurdle for a long time,” Ideker says. “But now it turns out that it is possible with the help of artificial intelligence.”

Left: Classic cell diagrams from the textbook imply that all parts are clearly visible and defined. Right: A new cell map generated by MuSIC reveals many new components. The yellow nodes represent known cell components, while the purple nodes represent the new components. The size of the node reflects the number of different proteins in the particular cell part. Image: UC San Diego Health Sciences

By merging multiple data and combining microscopy, biochemical techniques and artificial intelligence, researchers have now produced a better model of a cell. And that can really expand our understanding of human cells. The new technique is called Multi-Scale Integrated Cell (MuSIC) and uses deep learning to map a cell directly from cellular microscopy images. “The combination of technologies is unique and powerful,” said researcher Yue Qin. “It is the first time that measurements at vastly different scales have been brought together.”

Unknown Cell Parts

With the help of MuSIC, the researchers are now making some surprising discoveries. This technique thus reveals previously unknown cell components. In total, the researchers found about 70 components in a human kidney cell, half of which had never been seen before. And that means cells are home to dozens of parts we don’t know yet.

Strange structure

In addition, the team discovered a group of proteins that form an unknown structure. After further analysis, it appears that this strange structure is a new complex of proteins that binds RNA. The complex is likely involved in splicing, an important cellular event that helps determine which genes are activated at what time.

Model of a cell

The study is an important step forward. Because thanks to the research, scientists now have a more detailed model of a cell. However, MuSIC doesn’t map the cell contents to specific locations yet — like that aforementioned drawing from your biology textbook — partly because it isn’t necessarily fixed. Instead, the location of the components is “liquid” and can change depending on the cell type and situation.

According to Ideker, this study is just the beginning, looking only at 661 proteins and one cell type. “The obvious next step is to map the entire human cell,” he says. “Then we move on to different cell types, people and species. By then comparing the exact differences between healthy and diseased cells, we may be able to better understand the molecular basis of many diseases.”