Correcting gradient-based interpretations of deep neural networks for genomics

Genome Biol. 2023 May 9;24(1):109. doi: 10.1186/s13059-023-02956-3.

Abstract

Post hoc attribution methods can provide insights into the learned patterns from deep neural networks (DNNs) trained on high-throughput functional genomics data. However, in practice, their resultant attribution maps can be challenging to interpret due to spurious importance scores for seemingly arbitrary nucleotides. Here, we identify a previously overlooked attribution noise source that arises from how DNNs handle one-hot encoded DNA. We demonstrate this noise is pervasive across various genomic DNNs and introduce a statistical correction that effectively reduces it, leading to more reliable attribution maps. Our approach represents a promising step towards gaining meaningful insights from DNNs in regulatory genomics.

Keywords: Attribution methods; Deep learning; Explainable AI; Model interpretability; Regulatory genomics.

Publication types

  • Research Support, N.I.H., Extramural
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Genomics*
  • Learning*
  • Neural Networks, Computer
  • Nucleotides

Substances

  • Nucleotides