a | The transposable element content of the human genome. About 45% of the human genome can currently be recognized as being derived from transposable elements, the vast majority of which are non-LTR retrotransposons such as L1, Alu and SVA elements. L1, LINE-1; LTR, long terminal repeat. b | The canonical LINE-1 (L1) element consists of two open reading frames (ORF1 and ORF2) flanked by 5′ and 3′ untranslated regions (UTR). The 5′ UTR possesses an internal RNA polymerase II promoter (grey box). The element ends with an oligo dA-rich tail (AAA) preceded by a polyadenylation signal (pA). c | The canonical Alu element consists of two related monomers separated by an A-rich linker region (with consensus sequence A₅TACA₆). The left monomer contains A and B boxes (grey boxes) which are promoters of transcription by RNA polymerase III. The element ends with an oligo dA-rich tail (AAA) that can be up to one hundred bp in length. d | The canonical SVA element has a composite structure consisting of (from the 5′ end to 3′ end): (i) a (CCCTCT)_n hexamer repeat region, (ii) an Alu-like region consisting of two antisense Alu fragments and an additional sequence of unknown origin, (iii) a variable number of tandem repeats region made of 35-50 bp-long units, and (iv) a region derived from the env gene and the 3′ LTR of the endogenous retrovirus HERV-K10. The element ends with an oligo dA-rich tail (AAA) preceded by a polyadenylation signal (pA). L1, Alu and SVA elements are typically flanked by target site duplications (black arrows) generated upon integration. Elements are not drawn to scale.

a | Typical insertion of an L1, Alu or SVA retrotransposon (red box) at a new genomic site (blue area). If the new genomic site is a genic region, the retrotransposon may cause insertional mutagenesis. b | The protein products of an L1 element (red circle) may create DNA double-strand breaks (broken blue area). Alternatively, an existing double-strand break may be repaired via non-classical endonuclease-independent insertion of a retrotransposon (red box). c | Microsatellites (e.g. (TA)_n) may arise from the homopolymeric tracts endogenous to retrotransposons. d | Gene conversion may alter the sequence compositions of homologous retrotransposon copies (red and green boxes). e | Insertion of retrotransposons (red box) is sometimes associated with concomittant deletion of target genomic sequences (light blue box). f | Ectopic recombination (double arrowheaded line) between non-allelic homologous retrotransposons (red boxes) may result in genomic reaarangements such as deletions (left) or duplications (right) of intervening genomic sequences. g | 3′ and 5′ transduction can result in the co-retrotransposition of downstream or upstream flanking genomic sequence, respectively, along with a retrotransposon (left and right, respectively).

a | Retrotransposon sequence (red box) can be recruited as coding sequence and integrated into a gene (made up here of two exons, blue boxes). This is often associated with alternative splicing (dashed lines). b | Presence of a retrotransposon (red box) in the intron of a gene (sequence between the two blue boxes representing exons) can result in transcription elongation defects such as attenuation or premature termination. c | Retrotransposons (red box) carry transcription factor (green ovals) binding sites that can up- or down-regulate (green arrow) the expression of neighboring genes (horizontal arrow and blue boxes). d | Retrotransposons (red box) carry sense and anti-sense promoters (horizontal arrows) that can initiate downstream and upstream transcription. e | Presence of two Alu elements in opposite orientation (red boxes) in gene transcripts can lead to A-to-I editing, which can result in suppression of expression through nuclear retention of edited RNA transcripts. f | Retrotransposon sequences (red boxes) can be methylated, which may initiate and spread heterochromatin formation (green ovals), thereby altering expression of neighboring genes (horizontal arrow and blue boxes).