Google's AI genome system evaluates every possible one-base change
Most one-base changes to the human genome do nothing, but a few are significant.
Google unveiled AlphaGenome Atlas on Tuesday, a tool aimed at forecasting the effects of every conceivable single-base alteration within the human genome. The human genome spans approximately 3 billion bases, and to account for the remaining three DNA bases absent from the reference genome, AlphaGenome must process a total of 9 billion bases.
AlphaGenome's primary objective is to determine the potential functions of non-coding DNA. This type of DNA, which does not encode proteins, constitutes the majority of the human genome. However, certain areas of non-coding DNA play a crucial role in regulating the activity of protein-coding DNA. These regions direct the cell's actions regarding when and where to produce messenger RNAs and how to convert them into mature protein-coding forms.
Nonetheless, a significant portion of non-coding DNA appears to be remnants of ancient viruses and other molecular parasites. Identifying the functional elements within this vast expanse of DNA is beneficial, as having all the analysis performed by a single software package simplifies the process. Nonetheless, until biologists extensively utilize AlphaGenome (assuming they do), it remains uncertain what additional value this resource provides compared to its training data.
Written by urgent.news from Ars Technica's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
