Genetic Architect: Discovering Genomic Structure with Learned Neural Architectures
This addresses the challenge of designing deep learning architectures for genomics, where data structure is unknown, potentially benefiting researchers in computational biology and genetics.
The paper tackles the problem of decoding human genome structure by developing a novel search algorithm that learns optimal neural architectures from data, which simultaneously identifies key sequence motifs and predicts functional genomic outcomes. The resulting architectures surpass state-of-the-art results on benchmark genomics challenges using only RNA expression data.
Each human genome is a 3 billion base pair set of encoding instructions. Decoding the genome using deep learning fundamentally differs from most tasks, as we do not know the full structure of the data and therefore cannot design architectures to suit it. As such, architectures that fit the structure of genomics should be learned not prescribed. Here, we develop a novel search algorithm, applicable across domains, that discovers an optimal architecture which simultaneously learns general genomic patterns and identifies the most important sequence motifs in predicting functional genomic outcomes. The architectures we find using this algorithm succeed at using only RNA expression data to predict gene regulatory structure, learn human-interpretable visualizations of key sequence motifs, and surpass state-of-the-art results on benchmark genomics challenges.