Fig 1.
Transposable element (TE) distribution in 19 telomere-to-telomere genomes of Zymoseptoria tritici: (A) Origin of reference genome isolates originally used for the Z. tritici pangenome reported by Badet et al [52]. The color indicates the status of the dim2 gene with an important influence on RIP. Data from Möller et al [62]: dark blue = present and functional, bright blue = recently mutated, red = non-functional. Map created with ggmap version 3.0.0 [73]. Map data from OpenStreetMap: https://www.openstreetmap.org/copyright. (B) Genome size and TE copy number per isolate. Circle sizes indicate the genome size, the green shade indicates the TE content. The colors indicate MITEs (miniature inverted repeat transposable elements, small non-autonomous DNA transposons corresponding to several TE superfamilies), RLC and RLG (two superfamilies belonging to LTR) and LINE. (C) Copy numbers of TEs (left) and total length (right) in all 19 genomes. Smaller boxes correspond to TE families. (D) Allele frequency distribution of TEs at orthologous loci among genomes. TEs were defined as orthologous if they were located between the same set of orthologous genes.
Fig 2.
Characteristics of TE niches across the genome: (A) Proportional overlap of H3K27me2, H3K4me9 and H3K9me3 histone methylation marks with TE niches in the reference genome IPO323. Colors indicate the group of TE. (B) TE copy numbers between core and accessory chromosomes (copy number and density): in and outside the subtelomeric region (copies and density); into large RIP affected regions (LRAR; copy and density); into niches with a moderate (≥ 50%) or low (< 50%) GC content; overlapping regions annotated as genes.
Fig 3.
Distribution of niche and TE copy characteristics: (A) Overlap of TE content, gene content and GC content with TE copy niches. (B) Distribution of the distance to the closest gene in MITEs, RLC/RLG, LINE and other TEs. The red line indicates the mean distance. (C) Distances to the next TE MITEs, RLC/RLG, LINE and other TEs. The red line indicates the mean distance. (D) Pearson correlation matrix of 11 characteristics of TE copy niches and TE copy characteristics. Dark red indicates strong positive correlation, dark blue indicates strong negative correlation of two characteristics. * p < 0.05, ** p < 0.01, *** p < 0.001.
Fig 4.
Characteristics of high-copy TE families: TE families are ordered from the highest copy numbers to lowest copy numbers (right) in all 19 analyzed genomes combined. (A) Distribution total copy number per TE family and isolate. (B) GC content distribution per TE family. The red line represents a GC content of 50%. (C) Length of the consensus sequence corresponding to the full-length consensus sequence excluding nested TEs or partial deletions. (D) Nucleotide diversity of the TE family (transformed as log10(nucleotide diversity*100,000)). (E) Number of RIP-like mutation (CpA<->TpA/TpG<->TpA) per TE copy, corrected for the length of the TE. (F) Correlation between copy numbers and consensus sequence lengths for TE families. Circle size corresponds to the mean number of RIP-like mutations and the color indicates the nucleotide diversity.
Fig 5.
Genomic origin and features of TE transposition bursts: (A) Repeat landscape of the TE families with the highest copy numbers. Colors indicate the highest copy numbers (RII_Cassini, RLC_Deimos, RLG_Sol, RLG_Luna), TEs with multiple bursts (RLC_Deimos, RIX_Lucy and RLG_Luna) or very recent burst (Styx, Thrym). DTX_MITEs_Goblin is included but shows no apparent activity. (B) Normalized range of characteristics of TE copies and their genomic niche compared between the estimated ancestral states and derived copies. A positive value indicates that metrics increased compared to the ancestor sequence, and a negative value indicates a decrease. (C) Scheme of the definition of burst and burst outgroups based on phylogenetic trees of TE families. Green indicates the copies of a burst with low terminal branch lengths and the red outgroup indicates the closest related sister branch of a burst. (D) Distribution of niche and TE copy characteristics of copies belonging to a burst clade (blue) compared to all other copies (red).
Fig 6.
High copy number TE families and characteristics of burst initiation: Comparison of copies in bursts (red) and all other copies (blue) for the five TE families with the highest copy numbers. The TE families RLG_Luna and RLG_Sol have no copies assigned to burst clades (indicated by NA). The mutation rate was corrected for the length of the fragment and the number of copies across genomes.
Fig 7.
Phylogenetic reconstruction of the Deimos Copia retrotransposon: (A) Phylogenetic tree with colors indicating the number of RIP-like mutations. The black bar marks the different burst clades. The dot plot shows the changes in RIP-like mutations from the estimated ancestral state to the offspring for all internal and terminal branches from the ancestral state reconstruction. (B-E) Phylogenetic trees and ancestor-offspring changes for (B) the GC content of the niche, (C) the overlap of the niche with large RIP affected regions, (D) the GC content of the fragment and (E) the distance to the closest gene.
Fig 8.
Phylogenetic reconstruction of the Cassini retrotransposon: (A) Phylogenetic tree with colors indicating the number of RIP-like mutations. The black bar marks the different burst clades. The dot plot shows the changes in RIP-like mutations from the estimated ancestral state to the offspring for all internal and terminal branches based on the ancestral state reconstruction. (B-E) Ancestor-offspring changes for (B) the GC content of the niche, (C) the overlap of the niche with large RIP affected regions (LRAR), (D) the GC content of the fragment and (E) the distance to the closest gene.