Chromatin accessibility in the Drosophila embryo is determined by transcription factor pioneering and enhancer activation

Interpretable deep learning reveals the sequence rules of Hippo signaling

“…Indeed, the optimal spacing of the Tead double motif is not found more frequently than other spacings in the genome, e.g., a relative spacing of +2 bp is more frequent (Figure 4B). Moreover, BPNet has been shown to learn relative motif affinities 17,20,23,101 and thus might be adept at learning degenerate versions of a motif.…”

“…In this type of TF cooperativity, binding enhancement occurs when the motifs are spaced within ∼150 bp and are strongest the closer the motifs are. Such sequence rules point to nucleosome-mediated cooperativity 15,20,26,117–121 and may reflect the likelihood by which two motifs are covered by the same nucleosome. Although the mechanism is not well understood, it does not require specific interactions between TFs, which explains how signaling TFs can receive input from a wide variety of TFs in different cell types.…”

“…The key step is then to interrogate the model and extract the learned sequence rules using interpretation tools 15 . This reveals the learned TF motifs, including measurements of their relative affinities and the syntax rules, thus the distance relationships by which motifs cooperate with each other 15,17,20,22,23 .…”

“…On the other hand, strictly spaced motifs are not frequently observed in the genome, raising the question of whether TF binding cooperativity may also occur through more flexible motif syntax 27–29 . For example, we previously found that a TF may enhance the binding of another TF through soft motif syntax, which occurs at variable motif distances within ∼150 bp but is stronger at closer distances 15,20 .…”

“…To discover these potential sequence rules, we performed the TF binding experiments at the highest resolution and leveraged our previously developed deep learning model BPNet to predict the data at base resolution from genomic sequences of 1-kb 15,20,22,37,38 . This approach optimally resolves sequence rules between closely spaced motifs within enhancers 15 .…”