Additional data file 49.

This table shows blocks of modules that are adjacent to the ends of finished telomeres (see Materials and methods). The columns describe the same categories of information as indicated in Additional data file 48. A limited set of non-subtelomeric copies of subterminal duplicons exist (Additional data file 49). Their genomic locations suggest sites of ancestral telomere-associated chromosome rearrangements, including a well-documented telomere fusion at 2q13-q14 [37] that contains representatives of subterminal duplicon families A, B, C, and D (Additional data file 49). The non-subtelomeric site of a duplicon from family D at 3p12.3 is the tip of an extended duplication region; the DNA on the centromeric flank of this site contains 4q and 10q subtelomere homology, including beta satellite repeat structure resembling part of the D4Z4 repeat. Subterminal family F contains several non-subtelomeric sites of duplicons; those on chromosomes 22q, 14q, and 12p are very close to the respective centromeres (Additional data file 49), indicating potential ancestral inversion of a chromosome arm followed by duplication of pericentromeric sequences as a mechanism for the genesis of the non-subterminal copies of this subterminal sequence family. The sequence similarity between subterminal duplicon copies within a family is mainly in the 90-96% range for subterminal blocks A, B, and D (Table 2; see Additional data file 49 for the rare exceptions.). As with the subtel-only blocks, some of these duplicons correspond to only part of the subterminal block sequence. There is also some overlap in sequences occupied by subterminal duplicon blocks A, B, and D; this is reflected in their occupancy of parts of the same transcript families RPL23A7 and FAM41C (Table 2). The cross-family homologies between subterminal blocks A, B, and D are also in the 90-96% identity range but the positions of the duplicons within the blocks vary and are located at different distances from the (TTAGGG)n tract; also, there are several alternative organizations of high-copy repetitive elements (masked and not examined in detail in this study) within these subterminal blocks. Thus, there might be more frequent shuffling of subterminal sequences than sequences located more centromerically, at least within a subset of subtelomere alleles; this idea is broadly consistent with an earlier model of subtelomere structure featuring compartments with distinct functional properties [9]. Further refinement of the classification of these subterminal families appears feasible and will benefit from more extensive sampling of (TTAGGG)n-adjacent sequences from additional alleles. Subterminal Block F contains one duplicon on 10p with very high similarity to the 18p query sequence, suggesting a very recent duplication event; the remaining duplicons were all in the 91-94% identity range. Block C has the highest sequence similarity among all subterminal duplicon sequence families, and has a copy at the 2q fusion locus. Block E (96-97%) is unusual in that it corresponds to a portion of subtelomere-only duplicon family 6 (Table 1), and is the only subterminal duplicon sequence family with subtel-only properties. This particular sequenced allele of 17p might have formed by the truncation of a chromosome end within this large subtelomere-only duplicon, as there is mapping evidence for several longer alleles of the 17p telomere (H Riethman, unpublished). It is interesting to note that (TTAGGG)n tracts at 17p and, indeed, on this particular allele of 17p tend to be consistently among the shortest in the human genome [19,51].

Format: PDF Size: 44KB Download file

This file can be viewed with: Adobe Acrobat Reader

Ambrosini et al. Genome Biology 2007 8:R151   doi:10.1186/gb-2007-8-7-r151