SMiLe-seq specificities of transcription factors

Understanding the DNA binding properties of transcription factors (TFs) is crucial for the reverse engineering of gene regulatory networks from genomic data. Here we present a novel automated technology – SMiLe-seq – that allows us to determine the DNA binding specificities of TF monomers, homo- and heterodimers in a fast, robust and a cost-effective manner. The core of this technology is a microfluidic platform that performs selection of DNA specifically bound by TFs from a pool of randomized sequences. Coupled to high-throughput sequencing, this platform allows the characterization of TF DNA binding preferences at an unprecedented resolution. Unlike other, already established in vitro technologies that also aim to determine TF binding specificities, SMiLe-seq operates at micro scale and requires minute amounts of biological material, but produces specificity models that characterize even low-affinity and transient molecular interactions. Performing de novo motif discovery on SMiLe-seq data, we derived binding models for a number of TF monomers and heterodimers that generally agree with the TF binding models identified by ChIP-seq or by other comparable in vitro methods. In addition, for a number of factors, that were considered to be difficult to analyze by other methods, we uncovered binding motifs that were so far never reported thus expanding the repertoire of quantitative specificity models for TFs originating from various model organisms and humans.