NVIDIA And Harvard Use AI To Speed Up Genome Analysis

There are certain science experiments that can take a while before the results are returned. Take for example genome analysis. This is a process that can take a little over twos for it to be completed, but that could change in the future, thanks to the efforts of researchers at NVIDIA and Harvard.

The researchers have developed a deep-learning toolkit that will help speed up the process of genome analysis as well as lowering the costs of these tests and experiments. So far based on their initial run, the two day process we mentioned above has been shortened to just a mere 30 minutes, a fraction of the original time.

The genome analysis process, known as ATAC-seq, requires the analysis of tens of thousands of cells, whereas AtacWorks, the AI developed by the researchers, can get similar results with just tens of cells. The researchers also found that AtacWorks was capable of identifying separate parts of the DNA associated with white and red blood cells, something that typically cannot be studied using more traditional methods.

According to the NVIDIA’s lead researcher, Avantika Lal, “With very rare cell types, it’s not possible to study differences in their DNA using existing methods. AtacWorks can help not only drive down the cost of gathering chromatin accessibility data, but also open up new possibilities in drug discovery and diagnostics.”

Tyler Lee
Categories: Computers, Medical
Tags: AI (Artificial Intelligence), NVIDIA, Science