These sequence motifs which should bind HMGA2 as described by Cui and Leng 2007 have not been found in the DNA fragments revealed by ChIP. There are two possible explanations for the absence of corresponding sequences in the fragments identified by ChIP. First, HMGA2 proteins are not only interacting with DNA but also with a variety of other DNA binding and chromatin binding proteins like APEX1 Sorafenib and E4F1. If not distinguished from the highly similar HMGA1a and HMGA1b proteins, which is common in the literature, the HMGA proteins have a lot of molecular partners as transcription factors and other DNA binding proteins. Considering this facts together with the direct involvement of HMGA2 in base excision repair with own enzymatic lyase activity it is not likely that HMGA2 is binding DNA in a specific manner. Especially for the function in excision repair a specific binding site seems counterproductive to the occurrence of mutations only by chance not at specific sites. The second possible explanation for the absence of similarities between the SELEX sequences and the ChIP DNA fragments is,SP600125 that the occurrence of the consensus sequences for HMGA2 binding described by Cui and Leng is rare and the statistics of appearance of a 15 bp sequence implicates that these consensus sequences are of limited biological relevance for the HMGA2 activity. According to its function as a chromatin-remodelling switch HMGA2 is supposed to require a large number of DNA binding sites throughout the genome. This is consistent with the observations of relative abundance of HMGA2 in embryonic stem cells by Li et al.. Therefore, the rare occurrence of the consensus sequences has to be explained. Besides the possibility of artificial binding in the SELEX experiments the statistical 2.5 fold overrepresentation of the extended HMGA2 consensus sequences versus the representation of such sequences only by chance in the human genome points to a possible other explanation. The consensus motif described by Cui and Leng is efficiently binding HMGA2 but in vivo this binding is maybe irrelevant. The binding might be too strong for purposes of dynamic regulation which is required for the proper activity of HMGA2. The AT-content of the sequences generated by ChIP is significantly higher than the average of the human genome.