An AI that can predict cell structures 05-12

Fluorescent-labeled cells used to train neural networks. Image: Allen Institute. 


New 3D models of living human cells generated by machine-learning algorithms are allowing scientists to understand the structure and organization of a cell's components from simple microscope images.

Why it matters: The tool developed by the Allen Institute for Cell Science could be used to better understand how cancer and other diseases affect cells or how a cell develops and its structure changes — important information for regenerative medicine.

"Each cells has billions of molecules that, fortunately for us, are organized into dozens of structures and compartments that serve specialized functions that help cells operate," says Allen Institute's Graham Johnson, who helped develop the new model.

What they did: The researchers used gene editing to label the nucleus, mitochondria and other structures inside live human induced pluripotent stem cells (iPSC) with fluorescent tags and took tens of thousands of images of the cells.

They then used those images to train a type of neural network known as Generative Adversarial Networks (GANs). That yielded a model that can predict the most likely shape of the structures and where they are in cells based on just the cell's plasma membrane and nucleus.

Using a different algorithm, they created a model that can take an image of a cell that hasn't been fluorescent-labeled — in which it's difficult to distinguish the cell's components ("it looks like static on an old TV set," Graham Johnson says) — and find the structures.

What they found: When they compare the predicted image to actual labeled ones, the Allen Institute researchers said they are nearly indistinguishable.

The advance: Gene editing and fluorescent dyes often used to study cells only allow a few components to be visualized at once and can be toxic, limiting how long researchers can observe a cell.

Plus, "knowledge gained from more expensive techniques or ones that take a while to do and do well can be inexpensively applied to everyone’s data," says the Allen Institute's Greg Johnson, who also worked on the tool. "This provides an opportunity to democratize science."

View at the original source

Cellular division strategy shared across all domains of life 01-07

Archaea, bacteria, and eukarya use the same mechanism to maintain size...

SEAS researchers have found that these pink-hued archaea — called Halobacterium salinarum — use the same mechanisms to maintain size as bacteria and eukaryotic life, indicting that cellular division strategy may be shared across all domains of life. (Image courtesy of Alexandre Bison/Harvard University) 

The three domains of life — archaea, bacteria, and eukarya — may have more in common than previously thought.

Over the past several years, Ariel Amir, Assistant Professor in Applied Mathematics at the Harvard John A. Paulson School of Engineering and Applied Sciences (SEAS) has been studying how cells regulate size. In previous research, he and his collaborators found that E. coli (bacteria) and budding yeast (eukaryote) use the same cellular mechanisms to ensure uniform cell sizes within a population.

Now, with a team of collaborators including Ethan Garner, the John L. Loeb Associate Professor of the Natural Sciences at Harvard, and Amy Schmid, Assistant Professor of biology at Duke University, Amir found that archaea use the very same mechanism. 

The research is published in Nature Microbiology.

“These findings raise really interesting questions about how cellular mechanics evolved independently across all three domains of life,” said Amir. “Our results will serve as a useful foundation for, ultimately, understanding the molecular mechanisms and evolution of cell cycle control.”

Archaea are single-celled microorganisms that inhabit some of Earth’s most extreme environments, such as volcanic hot springs, oil wells and salt lakes. They are notoriously difficult to cultivate in a lab and, as such, are relatively understudied.

Archaea inhabit some of Earth’s most extreme environments, such as this salt lake in Bolivia (Image courtesy of Ariel Amir/Havard SEAS)

“Archaea are unique because they blend a lot of the characteristics of both bacteria and eukaryotes,” said Dr. Yejin Eun, first author of the paper. “Archaea resemble bacterial cells in size and shape but their cell cycle events — such as division and DNA replication — are a hybrid between eukaryotes and bacteria.”

The researchers studied Halobacterium salinarum, an extremophile that lives in high-salt environments. They found that like bacteria and budding yeast, H. salinarum controls its size by adding a constant volume between two events in the cell cycle. However, the researchers found that H. salinarum are not as precise as E.coli and there was more variability in cell division and growth than in bacterial cells. 

“This research is the first to quantify the cellular mechanics of size regulation in archaea,” said Amir. “This allows us to quantitatively explore how these mechanisms work, and build a model that explains the variability within the data and the correlations between key properties of the cell cycle. Eventually, we hope to understand just what makes this cellular mechanism so popular across all domains of life.” 

Archaea resemble bacterial cells in size and shape but their cell cycle events — such as division and DNA replication — are a hybrid between eukaryotes and bacteria.

View at the original source