Learning probabilistic protein–DNA recognition codes from DNA-binding specificities using structural mappings

Abstract

Knowledge of how proteins interact with DNA is essential for understanding gene regulation. Although DNA-binding specificities for thousands of transcription factors (TFs) have been determined, the specific amino acid–base interactions comprising their structural interfaces are largely unknown. This lack of resolution hampers attempts to leverage these data in order to predict specificities for uncharacterized TFs or TFs mutated in disease. Here we introduce recognition code learning via automated mapping of protein–DNA structural interfaces (rCLAMPS), a probabilistic approach that uses DNA-binding specificities for TFs from the same structural family to simultaneously infer both which nucleotide positions are contacted by particular amino acids within the TF as well as a recognition code that relates each base-contacting amino acid to nucleotide preferences at the DNA positions it contacts. We apply rCLAMPS to homeodomains, the second largest family of TFs in metazoans and show that it learns a highly effective recognition code that can predict de novo DNA-binding specificities for TFs. Furthermore, we show that the inferred amino acid–nucleotide contacts reveal whether and how nucleotide preferences at individual binding site positions are altered by mutations within TFs. Our approach is an important step toward automatically uncovering the determinants of protein–DNA specificity from large compendia of DNA-binding specificities and inferring the altered functionalities of TFs mutated in disease.

DeepPBS: Geometric deep learning for interpretable prediction of protein–DNA binding specificity

“…In this context, DeepPBS operates on these predicted complexes to yield the binding specificity of the system, which can guide improved modeling of protein-DNA complexes. DeepPBS, despite its generality, exhibits comparable performance to the recently described family-specific method rCLAMPS 28 .…”

“…3e-g). We show that this pipeline is competitive with the recent family-specific model rCLAMPS 28 (Fig. 3h-i) while being more generalizable: specifically, DeepPBS is family-agnostic, can handle biological assemblies, and can predict flanking preferences.…”

“…However, in conjunction with a complex structure prediction method, we can start from protein sequence information alone and predict binding specificity using DeepPBS. This process can be compared to the recent HD family-specific method, rCLAMPS 28 (see Supplementary Section 7). rCLAMPS can predict core 6mer specificities for monomer HD proteins.…”

Structure-based learning to model complex protein-DNA interactions and transcription-factor co-operativity incis-regulatory elements

“…In this work we have developed a structurebased approach to predict specific binding motifs of TFs, to identify cis-regulatory elements and to automatically model the structure of the transcription complex entailing the regulation. A similar structural-learning approach was recently developed and applied on homeodomain and C2H2-Zf families 63 that predicted the PWMs by mapping experimentally known PWMs in the contacts of the interface of the TF-DNA structure, which in principle (but not suggested) it can be applied to any other TF by providing the structure of the complex with DNA. Our approach has been implemented in a server for the scientific community named ModCRE.…”

Mining novel cis-regulatory elements from the emergent host Rhodosporidium toruloides using transcriptomic data

“…Nevertheless, new tools and methodological approaches are developing to predict regulatory elements in different datasets. In the future, they might help overcome motif discovery limitations in non-model organisms (Ksouri et al, 2021;Novakovsky et al, 2021;Wetzel et al, 2022).…”