Protein Atlas Maps the Inner Workings of Cells

The accumulation of misfolded proteins can cause disease. Knowing how proteins behave inside human cells can be a tool used to search for the origins of diseases such as dementia and many cancers. Now, researchers at the University of Cambridge led by Tuomas Knowles, PhD, professor of physical chemistry and biophysics proteins, have developed an atlas of proteins describing how they behave inside human cells. The atlas has allowed the researchers to discover new proteins inside cells that are responsible for a range of important bodily functions.

The atlas and the team’s findings are published in Nature Communications in an article titled, “Protein Condensate Atlas from predictive models of heteromolecular condensate composition.”

The team focused on a droplet-like part of the cell called a condensate which is a meeting hub for proteins to go and organize themselves. These hubs are also key sites where disease processes start.

The predictions are available with the paper so researchers around the globe can explore their protein targets of interest and any surrounding condensate systems.

“This model has allowed us to discover new components in membrane-less compartments in biology as well as discover new principles underlying their function,” said Knowles, who led the research.

“To date we have not had a comprehensive map of which proteins go together into which condensates, but in our work we provide a first such atlas,” said Knowles.

The rules directing proteins inside cells are not completely understood so the team decided to build this atlas to predict which proteins meet inside condensates.

“What motivated this research was the desire to understand the full complexity of protein condensates and to go a layer deeper than what scientists have researched so far,” said Kadi Liis Saar, PhD, first author on this research and a postdoctoral fellow at the Centre for Misfolding Diseases.

The researchers used large databases, such as StringDB and BioGRID, which contain data on many aspects of cells, along with more in-depth case studies about individual condensates.

“With this atlas, we can make predictions about every single protein in a cell, where exactly it would be found, and what sorts of other proteins it interacts with,” commented Saar. “We hope that this generates opportunities for researchers and opens up new possibilities for intervention in diseases associated with aberrant condensate formation.”

The AI found proteins present in the model cell that had never been observed before. If these proteins are now found in a lab then this is a good indicator that the AI is accurate.

“In our study, we discovered proteins within condensates that have never been seen there before. These proteins are involved in important functions in the body, such as the distribution of fat, the creation of actin inside cells, and the creation of new proteins. These proteins were not detected in the previous study that we used as our training set.

“We hope these data will enable new discoveries about the biological roles of condensates as well as the biophysical drivers behind condensate formation.”