Chromatin Landscapes of Retroviral and Transposon Integration Profiles
A) The bias of unselected integrations relative to CIS integrations, on a scale from blue (more CIS integrations) to red (more unselected integrations). B) log2 ratio of activating CISs and repressing CISs. A CIS is activating if it is not within a gene, or within a gene and 90% homogeneous with regard to orientation relative to that gene. Otherwise it is repressive. C) Bias of unselected integrations for CIS regions in a (i) genome-wide background, (ii) genic background (+/−100 kb), and (iii) intergenic background (whole genome except genes +/−100 kb), as measured by the log2 ratio of observed (unselected integrations) and expected (matched controls). D) CIS integration counts vs. unselected integration counts. CISs are annotated with the nearest TSS. Note that a single gene can be associated with multiple CISs. Spurious CISs were determined by a one-sided binomial test to determine if the CIS contained more CIS integrations than unselected integrations (, FDR-corrected).