NVIDIA and Harvard use Artificial Intelligence for fast and inexpensive genome analysis


Directing technology towards medicine allows it, in conjunction with science, to significantly improve the way it works in some fields. As we have seen, in recent years there have been several technological advances that represent clear advantages in health and in improving people's quality of life, especially involving Artificial Intelligence (AI). So researchers at Harvard and NVIDIA realized that using AI makes genome analysis faster and cheaper.

The combination of these investigations may soon bring technologies for the early detection of diseases that are currently critical for humanity.


Artificial intelligence can help in the detection of diseases such as cancer and Alzheimer's

Researchers in Harvard and NVIDIA developed a kit of tools deep-learning (learning deep) which is capable of significantly reducing the time and cost associated with conducting a rare and unicellular test. This technology could lead to the identification of biomarkers for diseases such as Alzheimer's or cancer.

Named AtacWorks, the kit can perform analysis of an entire genome, using NVIDIA Tensor Core GPUs. If this is normally a process for just over two days, this way it can be dispatched in half an hour.

The toolkit developed by Harvard and NVIDIA works with ATAC-seq, a method designed to find open areas in the genome of cells, both healthy and diseased. Open areas are understood to be subsections of a person's DNA that are used to determine and activate specific functions (for example, blood cells, liver, skin).

After all, this is the part of the human genome that can tell scientists whether or not a person has Alzheimer's, heart disease or cancer.


Nvidia technology allows analysis to require only dozens of cells

Typically, ATAC-seq requires the analysis of thousands of cells. However, AtacWorks, based on AI, can achieve the same results by analyzing only dozens.

In addition, Harvard and NVIDIA researchers applied a set of stem cell data that produces red and white blood cells. Although they are subtypes that cannot normally be studied using traditional methods, with AtacWorks they were able to identify separate parts of the DNA associated with white and red blood cells.

With very rare cell types, it is not possible to study differences in your DNA using existing methods. AtacWorks can help not only reduce the cost of collecting chromatin accessibility data, but also open up new possibilities in drug discovery and diagnosis.

Explained Avantika Lal, an NVIDIA researcher and lead author of the article.

With this new toolkit developed by Harvard and AI-based NVIDIA, scientists will be able to analyze the genome faster and cheaper. Thus, they will represent an important contribution in the identification of specific mutations or biomarkers that can cause diseases, such as cancer and Alzheimer's.

Post a Comment

0 Comments