Fig. 7

RepEnTools targeted search reveals enrichment of human UHRF1 Tandem-Tudor Domain binding on promoters of young, full-length L1PAs and the flanks of known enhancers. A The bar diagrams produced by RepEnTools, focusing on the L1 superfamily, reveal the subfamilies most and least enriched in TTD CIDOP versus input. The former contain the youngest evolutionary subfamilies (L1PA family), while the latter are dominated by the oldest (L1ME family). See also Additional file 1: Fig. S11A and Additional file 3: Table S1. B TTD enrichment is found at the 5′ end of many full-length L1Pas. This confirms RepEnTools’ findings, and is in agreement with hUHRF1 dependent L1 promoter DNA methylation and silencing of L1 transcription [43]. Heatmap of all L1PA models (pHMM), anchored to the 5′ end, and arranged by mean L1PA track intensity. L1PA track shows position and density of actual L1PA annotated segments within the model. pHMM – profile Hidden Markov Model. See also Additional file 1: Fig. S11B. C Restricting analysis to the full-length L1PAs reveals strong TTD enrichment in two reproducible peaks. Heatmap arranged by mean TTD track intensity. D hUHRF1-TTD CIDOP enrichment correlates well with the ratio of full-length elements within L1PA subfamilies (r 0.95). Only reproducibly enriched/depleted subfamilies are considered (p ≤ 0.05, n = 2). Enrichment of TTD over input was calculated by RepEnTools, while the ratio of full-length elements is shown in Additional file 1: Fig. S11C. See also Additional file 1: Fig. S11D. E TTD enrichment is highest around and on the L1PA antisense promoter (AS), overlapping the H3K4me1-K9me3 double mark. This is a known enhancer, specific to the L1PA lineage, that harbours transcription factor binding motifs [11, 45]. H3K4me1 and H3K9me3 ChIP-seq RPKM scale on the right axis [46]. The vertical axes are identical for all three profile plots of this figure. The change in scale reflects the differences in experimental and sequencing methods. RE track shows density of actual L1PA annotated segments within the model. RE annotation from dfam consensus sequence [42]. See also Additional file 1: Fig. S12A. F TTD enrichment on the youngest subfamily of L1PAs, L1HS/L1PA1, is centred on the antisense promoter (AS), experimentally shown to strongly enhance transcription of adjacent genes [47]. RE annotation from dfam consensus sequence [42]. See also Additional file 1: Fig. S12B. G TTD enrichment is strongest on the HERVK and LTR5-like regions of full-length SVAs, shown to harbour TEENhancers [13, 48], overlapping the H3K4me1-K9me2/3 double marks. Annotations compiled using comparative information from dfam [42]. See also Additional file 1: Fig. S12C-E. H CIDOP-qPCR experiments corroborated the TTD enrichments and depletions reported by RepEnTools using carefully designed and validated qPCR assays on selected targets. All data from n = 2 biological replicates, bar indicates mean. See also Additional file 1: Fig. S10E