Domain-adaptive neural networks improve cross-species prediction of transcription factor binding

The intrinsic DNA sequence preferences and cell type-specific cooperative partners of transcription factors (TFs) are typically highly conserved. Hence, despite the rapid evolutionary turnover of individual TF binding sites, predictive sequence models of cell type-specific genomic occupancy of a TF in one species should generalize to closely matched cell types in a related species. To assess the viability of cross-species TF binding prediction, we train neural networks to discriminate ChIP-seq peak locations from genomic background and evaluate their performance within and across species. Cross-species predictive performance is consistently worse than within-species performance, which we show is caused in part by species-specific repeats. To account for this domain shift, we use an augmented network architecture to automatically discourage learning of training species-specific sequence features. This domain adaptation approach corrects for prediction errors on species-specific repeats and improves overall cross-species model performance. Our results show that cross-species TF binding prediction is feasible when models account for domain shifts driven by species-specific repeats.

Automated summary

The DNA sequence preferences and cooperative partners of transcription factors (TFs) are conserved across species. However, predicting TF binding in one species based on sequence models of a closely related species is challenging due to species-specific repeats. To address this challenge, researchers used neural networks to predict TF binding across species and found that the predictive performance was worse than within-species predictions. By using an augmented network architecture, they were able to correct for prediction errors caused by species-specific repeats and improve the overall cross-species model performance.