TopTranscriptional regulation in eukaryotes
Supplementary material and resources from "Identification of sparsely distributed clusters of cis regulatory elements in sets of co-expressed genes" by Gabriel Kreiman
Supplemental figure 1
This figure illustrates how the number of distinct motif predictions from one of the motif finding algorithms that we used (alignACE, Hughes et al., 2000) decreases with the number of iterations. For each set of gene sequences, the alignACE algorithm was ran 10 times using the same input sequences and parameters. The x axis in the plots indicates the cumulative number of motif predictions. We compared each motif prediction with the previous set of motif predictions by considering the Spearman correlation coefficient as a measure of similarity between motifs (see Methods). The y axis indicates the cumulative number of non-redundant (using a threshold correlation coefficient of 0.75). The figure shows a saturation of the the number of novel predictions with subsequent iterations. (A) Set of random genes (see Performance on random sets of genes) (B) Set of human skeletal muscle genes (see Analysis of the muscle regulatory regions in humans) (C) Set of mouse homologs to the human skeletal muscle genes used in B (D) Set of fly pattern generation genes (see Analysis of the pattern generation set in flies)
