Extended Data Fig. 11: Enhancer validation experiments in human cell lines.

a) Schematic of the enhancer validation experiment flow. At top is the third-generation HIV-based self-inactivating vector (deletion in 3’ LTR indicated by red triangle), with PCR-amplified test DNA (blue, cloned in both orientations) inserted just 5’ of a basal Oct4 promoter (P) driving IRES-eGFP (green). Vector supernatant was prepared by plasmid co-transfection of 293T cells. Cells of interest were transduced and then analyzed by flow cytometry a few days later. Shown below is the expected post-transduction structure of the SIN HIV vector, with a duplication of the 3’ LTR deletion rendering both LTRs non-functional. b) Fold changes of gene expression of eGFP was compared between negative elements (n=20 biologically independent samples) and putative enhancers (n=20 biologically independent samples) chosen at random. Each sample in the plot is the average log fold change of the replicates for each element. c–e) Predicted enhancers increase gene expressions in A549, HOS, and TZM-bl cell lines. The enhancers were predicted in H1-hESCs. The activities of these enhancers (N=20 in each plot) were compared to control regions (N=20 in each plot) in three other cell lines: c) HOS, d) A549, and e) TZM-bl. The p-value were calculated by the two-sided t-test. The center value represented by the green line in the box plot shows the median log FC of each group. The 25th and 75th percentiles of the log fold changes in gene expressions for each group are represented by the upper and lower lines of the box, with whiskers connecting to the maximum and the minimum value.