AI is making an encyclopedia of human cells.
Our understanding of the structure of human cells, most of them from high school biology textbooks: cell membrane wrapped round nuclei, scattered around the center body, mitochondria, lysosomes and several other friends.
But scientists say that cell structure is not so simple at all.
Its composition is far more complicated than the purple onion under the microscope, and each part is in contact with each other. To figure out all this is very beneficial to doctors' diagnosis and disease mechanism research.
50 cell biologists, microscopic experts, and computer programmers from the Allen Institute, a cell science research center, recently published the Allen Integrated Cell Human Cell Research Tool.
Over the past two years, scientists have used genetic engineering to modify cell samples and label their internal structures (such as mitochondria) with fluorescence. They then provided photographs of more than 6,000 glowing cells to artificial intelligence algorithms to create a 3D model.
Now, in addition to the clear visibility of the 32,000 cell types in the database, this smart tool can also predict the internal operation of an unknown cell.
This cell tool focuses on human stem cells.
Greg Johnson, a scientist from the Allen Institute, believes that understanding the inner workings of healthy cells is very useful for the discovery and treatment of cancer. This means that doctors can “back to the past†from cancer cells and observe what has changed in the cells.
Rick Horwitz, director of the Allen Institute, also believes that cell evolution is more important than static morphology.
Even if you know the statistics of all players but never watch the game, you cannot predict the result of the football match.
He said that this tool may unexpectedly accelerate stem cell research, cancer research and drug development.
Optical microscopy is the most traditional method of cell observation. Although the cost is low and the operation is simple, its results are often frustrating.
You can see the clear contours of black and white. There are many small objects moving around and doing mysterious things. But you can't know what it's doing, and the lack of detail makes people as knowledgeable as high school students.
However, genetic modification, special dyes, or strong light irradiation often alter or damage the cells themselves, which is not conducive to long-term observation. Moreover, these technologies also rely on expensive equipment, which is difficult to achieve in some laboratories.
These are the reasons why you must collect images and let AI build 3D models.
Allen Integrated Cell's viewing page
Now with Allen Integrated Cell, you only need to provide a cell image under a microscope and artificial intelligence can restore its internal composition. This method does not lose details, nor is it costly and scary.
Susanne Rafelski, who is responsible for testing the R&D team, said:
It's like a photo of you with only a few wheels. It can also predict the brand and model of a car from limited information.
When only the fuzzy description is provided, as long as Allen Integrated Cell provides the cell size, shape, and location of the cell nucleus, artificial intelligence can also propose several different schemes to try to guess the structure and morphology of the entire cell.
Allen Integrated Cell artificial intelligence in two modes, one based on image restoration, the other is fuzzy guessing
Roger Brent, a researcher at the Fred Hutchinson Cancer Research Center, has been testing for several months. In his opinion, this tool has helped researchers around the world, including in less affluent areas.
Scientists at the Allen Institute stated that in the future they will also expand their research with the growth of stem cells, such as cardiomyocytes and kidney cells. They believe that this can further help biologists and doctors understand the development of certain diseases.
Brent is very confident about the future of this tool.
He believes that in 10 years, the mobile phone will become powerful enough to carry this tool. This may mean that even in developing countries, a small laboratory can judge and select better treatment options from the inside of cancer cells.
xcool vape enjoy free, Disposable Vape Pen brand
Xcool vapor disposable vape pen, xcool vapor hnb, xcool vapor cbd
Shenzhen Xcool Vapor Technology Co.,Ltd , https://www.szxcoolvape.com