Recent segmental and gene duplications in the mouse genome

Abstract

Background: The high quality of the mouse genome draft sequence and its associated annotations are an invaluable biological resource. Identifying recent duplications in the mouse genome, especially in regions containing genes, may highlight important events in recent murine evolution. In addition, detecting recent sequence duplications can reveal potentially problematic regions of the genome assembly. We use BLAST-based computational heuristics to identify large (>/= 5 kb) and recent (>/= 90% sequence identity) segmental duplications in the mouse genome sequence. Here we present a database of recently duplicated regions of the mouse genome found in the mouse genome sequencing consortium (MGSC) February 2002 and February 2003 assemblies.

Results: We determined that 33.6 Mb of 2,695 Mb (1.2%) of sequence from the February 2003 mouse genome sequence assembly is involved in recent segmental duplications, which is less than that observed in the human genome (around 3.5-5%). From this dataset, 8.9 Mb (26%) of the duplication content consisted of 'unmapped' chromosome sequence. Moreover, we suspect that an additional 18.5 Mb of sequence is involved in duplication artifacts arising from sequence misassignment errors in this genome assembly. By searching for genes that are located within these regions, we identified 675 genes that mapped to duplicated regions of the mouse genome. Sixteen of these genes appear to have been duplicated independently in the human genome. From our dataset we further characterized a 42 kb recent segmental duplication of Mater, a maternal-effect gene essential for embryogenesis in mice.

Conclusion: Our results provide an initial analysis of the recently duplicated sequence and gene content of the mouse genome. Many of these duplicated loci, as well as regions identified to be involved in potential sequence misassignment errors, will require further mapping and sequencing to achieve accuracy. A Genome Browser database was set up to display the identified duplication content presented in this work. This data will also be relevant to the growing number of investigators who use the draft genome sequence for experimental design and analysis.

Figure 1

Intrachromosomal segmental duplications identified in the mouse genome (chromosomes 1-X; results are based on the February 2003 assembly). Each line represents a duplicated module and connects a paralogous duplicon pair. Red, 99-100% sequence identity; purple, 96-98%; green, 93-95%; and blue, 90-92%. Correspondences to chromosome ideograms (obtained from Ensembl) are only crude. Graphics were produced using GenomePixelizer [34].

Figure 2

The genomic organization of the Mater duplication. (a) Location of the Mater duplication. A snapshot view of GMOD browser (details can be found at [14]). (b) Chromosomal view (mouse chromosome 7) of the three Mater duplication locations (DUP1, DUP2, MaterP). (c) Graphical view of the sequence similarity between DUP1 and DUP2 shown by GenomePixelizer. DUP2 is situated in an inverse orientation with respect to DUP1. Red, 99-100% sequence identity; purple, 96-98%; green, 93-95%; blue, 90-92%; black, 85-89%. (d) Graphical view of the sequence similarity between DUP1 and the MaterP region. As shown, MaterP is an intron-less, retrotransposed pseudogene. Blue, 90-92% sequence identity; black, 85-89%.

Figure 3

FISH detection of Mater duplication. (a) Metaphase FISH showing three pairs of signals (yellow) detected on mouse chromosome 7 using BAC clone RP23-225F5 (detection frequency of 70%) mapping to duplicated Mater regions. (b) DAPI banding of the same partial mitotic figures for the identification of mouse chromosome 7. A control probe RP23-464L20 was mapped to a single location in the F2 region (data not shown).