Finally, machine learning interprets gene regulation clearly | Molecular & Computational biology - Phys.org

In this age of "big data," artificial intelligence (AI) has become a valuable ally for scientists, according to Brian Stallard, Cold Spring Harbor Laboratory.

Assistant Professor Justin Kinney showcases the relatively easy-to-understand structure of a newly-designed artificial neural network. His results were officially presented at the 1st Conference on Machine Learning in Computational Biology on December 13.

Machine learning algorithms, for instance, are helping biologists make sense of the dizzying number of molecular signals that control how genes function. But as new algorithms are developed to analyze even more data, they also become more complex and more difficult to interpret. Quantitative biologists Justin B. Kinney and Ammar Tareen have a strategy to design advanced machine learning algorithms that are easier for biologists to understand. 

The algorithms are a type of artificial neural network (ANN). Inspired by the way neurons connect and branch in the brain, ANNs are the computational foundations for advanced machine learning. And despite their name, ANNs are not exclusively used to study brains.
Biologists, like Tareen and Kinney, use ANNs to analyze data from an experimental method called a "massively parallel reporter assay" (MPRA) which investigates DNA. Using this data, quantitative biologists can make ANNs that predict which molecules control specific genes in a process called gene regulation...

Now, Kinney and Tareen developed a new approach that bridges the gap between computational tools and how biologists think. They created custom ANNs that mathematically reflect common concepts in biology concerning genes and the molecules that control them. In this way, the pair are essentially forcing their machine learning algorithms to process data in a way that a biologist can understand.
Read more... 

Additional resources 
Ammar Tareen et al. Biophysical models of cis-regulation as interpretable neural networks, bioRxiv (2019).  
DOI: 10.1101/835942 

Source: Phys.org