Figure 1.
Overview of different approaches for identifying orthologous regions in two genomes.
Sequence-based methods (e.g. ENSEMBL Compara) start with short local alignments that are extended to longest possible alignments over gaps. Breakpoint-based methods use orthologous elements (called ‘anchors') to find the minimum number of rearrangements that transforms one genome into the other. Positional orthology tries to distinguish orthologs from paralogs by analyzing gene neighborhoods.
Figure 2.
Detection of collinear blocks by SyntenyMapper.
A: Illustration of a syntenic region between two species, with numbered boxes representing genes and connecting lines representing orthology relationships. Gene and gene
have no orthologs in their syntenic regions, but are orthologous to each other. Genes
and
have no orthologs. B: During pre-processing one-to-many (genes
and
,
) and asymmetric many-to-many (genes
and
), groups are first converted into symmetric groups by excluding genes with the lowest sequence identity to the rest of the group (genes
and
), and subsequently paired as one-to-one orthologs based on gene order. Breakpoints (zig-zag lines) are identified as described in the Methods section. C: Using breakpoints, SyntenyMapper defines rearranged segments, shown in black, as new syntenic regions 1_1 to 1_3 within the long original region.
Figure 3.
Breakpoint definition in SyntenyMapper.
Illustration of two breakpoints emerging at both ends of a translocated segment in genome A and
in genome B (hatched box). By definition a breakpoint is constituted by two orthologous gene pairs
and
if
, as shown in the boxes underneath the schema. The second breakpoint is described by
and
. White and black boxes mark the four genes forming the first
and the second
breakpoint, respectively. A is used as reference genome to define the block formed by a micro-rearrangement as the genes that lie between the adjacent breakpoints in A, in this case
,
and
. The genes between these two breakpoints and their orthologs in B form a block.
Table 1.
Frequency of different cases of orthologous relationships for a given gene in a syntenic region between human and mouse.
Figure 4.
Visualization of SyntenyMapper results for a syntenic region in human and mouse.
The region (ENSEMBL identifier 44542) is illustrated in human as a dark grey ideogram (right) and in mouse as a light grey ideogram (left). Ticks are placed at 100 kB distance and the numbers represent positions in mB on the human and mouse chromosomes 15 and 7, respectively. The Circos circular plot illustrates the positions of genes/intergenic regions for one syntenic region in both species and the correspondence between them. Micro-rearrangements are illustrated by color-coding, with syntenic orthologs and out-of-order genes shown in grey and black, respectively, while the intergenic regions between syntenic orthologs and between out-of-order genes are shown in white. A large block of seven genes (black) was translocated in either human or mouse. In the Galaxy version of the plots, gene annotations are given as labels and as direct links to ENSEMBL through clicks onto the gene track.
Figure 5.
Relationship of syntenic region length, evolutionary distance and other features.
Dependence of synteny features on the average syntenic region length (x axis) and evolutionary distance (circle size, inferred from branch lengths in Miller et al (2007), calculated as the average number of substitutions per site). Crosses correspond to the species pairs with no distance information available. A) Negative exponential correlation between the number of syntenic regions and their average length (inset: logarithmic axis scales). Closely related species (small circles) tend to have fewer, longer syntenic regions. Distant species (large circles) tend to have high numbers of very short syntenic regions. B) The average number of internal micro-rearrangements per syntenic region strongly correlates with syntenic region length and evolutionary distance. Evolutionary distant pairs of species share short syntenic regions with few internal micro-rearrangements. C) No clear correlation between syntenic region length and rearrangement number can be observed for external micro-rearrangements. D) The size of internal micro-rearrangements (average number of genes involved) does not correlate with the syntenic region length and evolutionary distance. E) Similarly, the number of genes involved in external translocations is generally independent of the syntenic region length. However, with the exception of mouse and platypus (average number of genes in externally translocated regions: 1.0, marked by arrow), distant species pairs tend to have somewhat longer externally translocated regions (average>1.05).