ShapeME: A Tool and Web Front-end for De Novo Discovery of Structural Motifs Underpinning Protein-DNA Interactions.

Determining where proteins bind a genome is paramount to understanding gene regulation. In addition to DNA sequence motifs, structural motifs (e.g., a narrow minor groove width) determine binding for some proteins (Rohs et al., 2009) [1]. Algorithms using structural features of DNA to predict protein binding exist (Mathelier et al., 2016; Samee et al., 2019; Yang et al., 2019; Pal et al., 2019) [2-5], but a structural motif discovery framework which can be applied to a variety of experimental designs is needed. We present a workflow capable of utilizing virtually any type of data representing sequence coverage or enrichment (e.g. ChIP-seq, RNA-seq, SELEX, etc.) to discover structural motifs with explanatory power for a protein's DNA binding preference. Our approach to motif discovery wraps shape and sequence motif inference into a single tool called ShapeME (github: https://github.com/freddolino-lab/ShapeME.git, web interface: https://seq2fun.dcmb.med.umich.edu/shapeme). Application of ShapeME to ENCODE datasets reveals proteins for which short structural motifs outperform the best PWM for that protein at the JASPAR database, or as identified by the sequence motif elicitation tool STREME. ShapeME is a powerful, versatile framework for inferring structural DNA binding motifs.