The role of L1 retrotransposition in human genomic variability / Adam D. Ewing.

Format:

Book

Manuscript

Thesis/Dissertation

Author/Creator:

Ewing, Adam D.

Contributor:

Kazazian, Haig H., Jr., advisor.

University of Pennsylvania.

Language:

English

Subjects (All):

Penn dissertations--Genomics and computational biology.

Genomics and computational biology--Penn dissertations.

Academic Dissertations as Topic.

Medical Subjects:

Academic Dissertations as Topic.

Local Subjects:

Penn dissertations--Genomics and computational biology.

Genomics and computational biology--Penn dissertations.

Physical Description:

ix, 216 pages : illustrations (some color) ; 29 cm

Production:

2010.

Summary:

LINE-1 (L1) elements are autonomous retrotransposons that replicate through a copy-and-paste mechanism, and constitute about 17% of the human genome. They are actively retrotransposing new copies from a human-specific subfamily, thus contributing to inter-individual genomic variation. We developed a technique to assay this L1-induced variation on a large scale using targeted high-throughput sequencing. In doing so, we find that any two individuals differ at an average of 285 dimorphic insertion sites with respect to presence or absence of an L1. Using data generated by this technique across 15 unrelated human genomes, we estimate the rate of retrotransposition at approximately 1 new insertion per 140 live births and the number of common insertion alleles in the human population at 3,000 to 10,000.

Additionally, we have developed computational methods to identify L1 insertions from whole-genome resequencing data such as the 1000 Genomes Project. An analysis of these data yielded 953 insertions present in one or more individuals but absent from the reference genome. The scale of this dataset allows us to identify L1 insertions specific to various populations, and we observe that the number of insertions specific to African populations outnumbers the insertions specific to other groups by a factor of 26. Cross referencing this set of insertions with our previous L1-targeted resequencing results, and with data from a number of other studies, yields over 1600 potential insertions not represented in the human reference genome assembly, 1018 of which have been independently validated.

Our L1-targeted resequencing technique allows us to detect de novo L1 insertions from endogenous L1 elements, which was previously only possible using engineered L1 retrotransposons carrying marker cassettes. We demonstrate this by identifying seven clone-specific endogenously-generated insertions in the genomic DNA of clonallyderived embryonic carcinoma and embryonic stem cell lines. We also show that our technique can identify de novo retrotransposition events in vivo with the identification of an insertion in genomic DNA from fetal placental tissue that is not present in the genomes of either parent. In summary, L1 insertion polymorphisms are a widespread phenomenon actively contributing to the genomic diversity of human beings.

Notes:

Adviser: Haig H. Kazazian, Jr.

Thesis (Ph.D. in Genomics and Computational Biology) -- University of Pennsylvania, 2010.

Includes bibliographical references.