Characterization of Human Endogenous Retrovirus sequences identifed in the human genome using the RetroTector software

Human Endogenous Retroviruses (HERVs) represent the inheritance of ancient germ-line cell infections by exogenous retroviruses and the subsequent transmission of the proviral integrated elements to the descendants. Actually, no replication-competent HERV sequence is recognizable in the human genome. However, some HERVs retain one or several intact retroviral genes and may express protein products that could interfere with the human immune system. The number and classification of HERVs vary according to method of enumeration. The focus of this project is to perform a systematic analysis and a classification of the most intact HERV sequences in order to better understand their evolution and their involvement in shaping the human genome. The human genome assembly GRCh 37/hg19 was analyzed with RetroTector software and a total of 3290 HERV proviral sequences were identified. The complex genetic structure of the 3290 proviruses was resolved through a multi-step classification procedure that involved a novel type of similarity image analysis (Simage). The 3290 HERV were classified in 40 unique clades (groups) which could be placed into class I (Gamma- and Epsilon-like), II (Beta-like) and III (Spuma-like). Simage analysis contributed to define the presence of a high number (around 40%) of mosaic forms, with heterogenous sequence content. A finest characterization of the HERV sequences was achieved with the investigation of a broad panel of structural markers that contributed to confirm and extend the previously performed classification. Finally, the HERV background of integration was also studied. Integration patterns analysis showed a tendency for proviruses from the same clade to occur together, within 100000 bases, maybe due to local duplications. Representatives from some gammaretroviral clades (HERVH and HERVE) integrated more frequently than expected by chance into the 5´end of transcriptional units, mostly in antisense orientation. A few lncRNAs were also found to contain HERV sequences. Thus, cis-effects from HERVs are to be expected. In conclusion, this study represents an advance in the state-of-the art of HERV characterization within the human genome and a starting point for upcoming studies on HERVs.