PMC

The mean and the standard deviation of similarities between TFBS models for the same TF. Similarities are computed for HOCOMOCO and JASPAR TFBS models for 85 TFs.

The circular tree illustrating the hierarchy of high quality models from HOCOMOCO collection. Clusters are shown by alternating colors. The examples of clustered TFBS models are shown with respective LOGO representations. The tree is drawn using jsPhyloSVG [].

The cumulative distributions (a) and probability density (b) of similarities for pairs of TFBS models. The similarities for pairs of models for the same TF are shown by solid lines (data for 85 TFs with the models available in both HOCOMOCO [] and JASPAR [] databases). The similarities for all possible pairs for 170 assessed models are shown by dashed lines. Different colors correspond to different P-value levels. It is notable that the paired models for the same TF are really closer as compared with the whole set of possible pairs.

The similarities (depending on P-value) and LOGO representations for pairs of TFBS models (HOCOMOCO and JASPAR) for selected TFs. It is notable that even for extremely similar LOGOs, like those of CTCF, the Jaccard similarity reaches only 0.6, indicating that the models define the sets of binding sites overlapping only for 60%. The similarity remains comparatively low even at high P-values (e.g. 0.01 where each 100^th word of the dictionary is recognized as the binding site). The same effect is shown for KLF4 (with the exception of similarity 1.0 for the lowest P-value, where both models recognize only identical consensus sequences). SPI1 models differing in length show very weak similarities. HIF1A models are surprisingly dissimilar at low P-values (possibly due to shorter model lengths).