Epistatic Net allows the sparse spectral regularization of deep neural networks for inferring fitness functions

Amirali Aghazadeh, Hunter Nisonoff, Orhan Ocal, David H. Brookes, Yijie Huang, O. Ozan Koyluoglu, Jennifer Listgarten, Kannan Ramchandran

Research output: Contribution to journalArticlepeer-review

18 Scopus citations


Despite recent advances in high-throughput combinatorial mutagenesis assays, the number of labeled sequences available to predict molecular functions has remained small for the vastness of the sequence space combined with the ruggedness of many fitness functions. While deep neural networks (DNNs) can capture high-order epistatic interactions among the mutational sites, they tend to overfit to the small number of labeled sequences available for training. Here, we developed Epistatic Net (EN), a method for spectral regularization of DNNs that exploits evidence that epistatic interactions in many fitness functions are sparse. We built a scalable extension of EN, usable for larger sequences, which enables spectral regularization using fast sparse recovery algorithms informed by coding theory. Results on several biological landscapes show that EN consistently improves the prediction accuracy of DNNs and enables them to outperform competing models which assume other priors. EN estimates the higher-order epistatic interactions of DNNs trained on massive sequence spaces-a computational problem that otherwise takes years to solve.

Original languageEnglish (US)
Article number5225
JournalNature communications
Issue number1
StatePublished - Dec 1 2021

ASJC Scopus subject areas

  • General Chemistry
  • General Biochemistry, Genetics and Molecular Biology
  • General Physics and Astronomy


Dive into the research topics of 'Epistatic Net allows the sparse spectral regularization of deep neural networks for inferring fitness functions'. Together they form a unique fingerprint.

Cite this