In an amazing paper published in Science Express, Bulyk, Hughes and colleagues haved used protein-binding microarrays to determine the DNA-binding specificities of 104 mouse transcription factors.
The most interesting finding of the study is that nearly half of all TF seem to be able to bind two distinct types of sequences. For example, Hnf4a appears to bind to both GGTCA and GGTCCA (and this was confirmed by EMSA). This is also the case for BCL6, which I actively study. They also found several clear cases of dependencies between motif positions (not only adjacent positions).
This study provides a very rich dataset that will be extremely useful for those of us who work on transcriptional regulation.
