Overview of the ArchAlign algorithm. (a) High-resolution chromatin structural data from regions of interest are converted into log2 ratios and aligned within a user defined alignment frame. (b) ArchAlign has two options for aligning chromatin structural patterns: 1, single-best-pair uses the two regions with the highest similarity, by Pearson or Spearman correlation or by Euclidian distance, to seed a single progressive alignment; 2, seed sampling is a more comprehensive search of the possible alignment space, which uses every region within the alignment as a possible seed for an independent alignment. The best alignment is selected as the one that maximizes the correlation across all regions in the alignment frame.
Lai and Buck Genome Biology 2010 11:R126 doi:10.1186/gb-2010-11-12-r126