Web supplement to
"Diverse CGG binding proteins across eukaryotes have been produced by independent domestications of hAT transposons"

The human transcription factor (TF) CGGBP1 (“CGG Binding Protein”) is conserved only in amniotes, and is believed to derive from the zf-BED and Hermes transposase DNA-binding domains (DBDs) of a hAT DNA transposon. Here, we show that TFs with this bipartite domain structure have resulted from dozens of independent hAT domestications in different eukaryotic lineages. CGGBPs display a wide range of sequence specificity, usually including preferences for CGG or CGC trinucleotides, while some bind AT-rich motifs. The CGGBPs are almost entirely non-syntenic, and their protein sequences, DNA binding motifs, and patterns of presence or absence in genomes are uncharacteristic of ancestry via speciation. At least eight CGGBPs in the coelacanth Latimeria chalumnae bind distinct motifs, and the expression of the corresponding genes varies considerably across tissues, indicating two overlapping modes of neofunctionalization.

• Web Supplementary File 1:File summarizing Protein Binding Microarray assays, including Construct IDs, gene names, protein accessions and source database, DNA sequences, and AA sequences. (excel file)
• Web Supplementary File 2:Weighted PFMs produced using Top10AlignZ. PFMs are named according to the corresponding Construct ID. (txt file)
• Web Supplementary File 3:PBM plasmid backbone maps. (doc file)
• Web Supplementary File 4:Forward and reverse logos for all weighted PFMs in Supplementary File 2. (zip file)

• Figure 1. Three tables containing E scores from Figure 1, with assays as columns, and 8-mers in rows. Includes unscaled E scores, scaled E scores (10^(E*10)-3)) and E scores and that have been scaled and normalized between 0-100 (the values in the heatmap itself).(zip file)
• Figure 2. Source alignment for the phylogram in Figure 2, as well as the alignment without trimmed columns. Also includes the phylogram with bootstrap support values as branch labels, in Raxml Newick format.(zip file)
• Figure S1. Percent identity matrix underlying heatmap in Supplementary Figure S1, the source multiple sequence alignment, and annotations representing which proteins were assayed on PBMs.(zip file)
• Figure S2. Source amino acid multiple sequence alignment for Supplementary Figure S2.(zip file)
• Figure S5. CGGBP expression data tables for Supplementary Figure S5 heatmap, in multiple formats. Columns are different RNA-seq runs (each corresponding to an SRA run), and rows are each coelacanth CGGBP gene. Includes table of log2(TPM + 1) values, which are the values in the heatmap, as well as a table of TPM values, and table of raw counts. Also includes table of SRA runs analyzed, and relevant metadata. All SRA runs were part of Nikaido et al., 2013.(zip file)

PBM Data are available at GEO accession GSE157085. HK and ME array designs are available at GEO accession GPL11260.