A study published in Science (2009) reveals the complex and diverse DNA-binding specificities of 104 distinct mouse transcription factors (TFs) representing 22 structural classes. Using a universal protein binding microarray (PBM) technology, the researchers determined the DNA-binding preferences of these TFs, finding that nearly every protein has unique binding preferences. Many TFs recognize multiple distinct sequence motifs, challenging previous assumptions about how proteins interact with DNA. The study highlights the importance of low-affinity binding sites in gene regulation and evolution of transcriptional regulatory networks. The data show that TFs can recognize a wide range of DNA sequences, with some exhibiting position interdependence and multiple binding modes. For example, the TF Jundm2 binds to a 'variable spacer length' motif, while others use combinations of position interdependence and variable distances between parts of their motifs. The researchers also identified secondary motifs that represent alternative binding preferences not captured by primary motifs. These secondary motifs can be used to predict TF binding sites and may explain differences in TF function. The study found that some TFs recognize their DNA binding sites through multiple completely different interaction modes, suggesting that TFs may use alternate structural features or conformations to bind DNA. The data suggest that TFs bind a rich spectrum of k-mers, and that multiple motif models are more accurate than single motif models for representing TF binding profiles. The study also highlights the importance of considering the quantitative nature of k-mer binding data in scoring candidate regulatory elements. The findings have important implications for understanding how proteins interact with their DNA binding sites and for genome analysis. The data are likely to be highly informative for well-conserved homologs in other organisms, and generating PBM data for all regulatory factors in major model organisms is an important goal.A study published in Science (2009) reveals the complex and diverse DNA-binding specificities of 104 distinct mouse transcription factors (TFs) representing 22 structural classes. Using a universal protein binding microarray (PBM) technology, the researchers determined the DNA-binding preferences of these TFs, finding that nearly every protein has unique binding preferences. Many TFs recognize multiple distinct sequence motifs, challenging previous assumptions about how proteins interact with DNA. The study highlights the importance of low-affinity binding sites in gene regulation and evolution of transcriptional regulatory networks. The data show that TFs can recognize a wide range of DNA sequences, with some exhibiting position interdependence and multiple binding modes. For example, the TF Jundm2 binds to a 'variable spacer length' motif, while others use combinations of position interdependence and variable distances between parts of their motifs. The researchers also identified secondary motifs that represent alternative binding preferences not captured by primary motifs. These secondary motifs can be used to predict TF binding sites and may explain differences in TF function. The study found that some TFs recognize their DNA binding sites through multiple completely different interaction modes, suggesting that TFs may use alternate structural features or conformations to bind DNA. The data suggest that TFs bind a rich spectrum of k-mers, and that multiple motif models are more accurate than single motif models for representing TF binding profiles. The study also highlights the importance of considering the quantitative nature of k-mer binding data in scoring candidate regulatory elements. The findings have important implications for understanding how proteins interact with their DNA binding sites and for genome analysis. The data are likely to be highly informative for well-conserved homologs in other organisms, and generating PBM data for all regulatory factors in major model organisms is an important goal.