Figures and data
Characterization of rice TIP-eQTLs.
A-B, Association between TIPs and gene expression levels in cis (TIP-eQTLs) found in indica and japonica population along the twelve rice chromosomes. Horizontal lines represent 5% significance thresholds corrected for multiple testing using FDR method. C, Number of TIP-eQTLs found in each condition classified at the order TE level. White bars represent the number of total TIPs (secondary axis). D, Expression variance explained by leading TIP-eQTL and SNP-eQTL. Each point represents a gene. Indica and Japonica TIP-eQTLs are combined into a single plot. Number of TIP-eQTLs (E) or SNP-eQTL (F) per Mb of gene feature. Upstream and downstream regions are 1 Kb long. Gene region includes 5’ and 3’ UTRs. All the TIPs and SNPs with a significant association (FDR < 5%) were used in this analysis.
Effect size and population frequencies of wet and drought-stress TIP-eQTLs.
Representation of the effect size (beta) of the TIP-eQTLs with respect to their population frequency in indica (A, C) and japonica (B,D) for positive (A,B) and negative (C,D) effects. Venn diagram illustrating the intersection between TIP-eQTLs detected in each condition of the two subspecies analyzed, indica (ind), and japonica (jap) (E). Percentage of condition-specific (F) and species-specific (G) TIP-eQTLs. Shared TIPs are those falling in the intersection between the FDR-corrected TIP-eQTLs found in a given population/condition and all the associations of the other population/condition using a relaxed cutoff (p < 0.05, no FDR correction). Relationship between the frequencies in indica and japonica populations of the and 829 TIP-eQTLs (H) and 33,389 non-eQTL TIPs (I).
Signatures of positive selection on TIP-eQTLs.
A, Absolute PBS of 354 TIP-eQTLs (left) and 11,344 TIPs (no-eQTL, right) present in indica, japonica and rufipogon populations. Dotted vertical lines represent the 95th an 99th percentile of the PBS values of the whole dataset (11,698 TIPs). Red dots represent two examples of TIP-eQTLs with extreme PBS values. B) Population frequency of the two TIPs with extreme PBS values, marked as red dots in the left panel A, as well as of the whole TIP dataset. C) Fst-based tree of the two TIPs with extreme PBS values, as well as of the whole TIP dataset (average Fst).
Selection on TIP-eQTLs associated with EG2 expression.
A, Representation of the two main EG2 haplotypes present in rice and rufipogon/nivara populations identified in the rice super-pangenome. Conserved nucleotide regions are connected by gray marks. Structural variants longer than 50bp are shown as white spaces. TIP-eQTLs are shown as red boxes. Additional TIPs are shown as white boxes. B, Boxplot representation of the expression of the two EG2 haplotypes in the indica population. Numbers inside boxplots represent the number of accessions in each group. Analysis of nucleotide diversity (π) and diversity in haplotypes homozygosity tracts (H12) for Hap A (C, E) and Hap B (D, F) in japonica (C, D) and indica (E, F) populations. The vertical dotted black line shows the position of the TIP insertion. the EG2 gene is schematically shown in blue. The horizontal dashed black line represents the mean of 1000 random permutation pulls (for p-value see Supplementary file 6).
Selection on TIP-eQTLs associated with OsGAP expression.
A, Representation of the two main OsGAP haplotypes present in rice and rufipogon/nivara populations identified in the rice super-pangenome. Conserved nucleotide regions are connected by gray marks. Structural variants longer than 50bp are shown as white spaces. TIP-eQTLs are shown as red boxes. Additional TIPs are shown as white boxes. B, Boxplot representation of the expression of the two OsGAP haplotypes in the japonica population. Numbers inside boxplots represent the number of accessions in each group. Analysis of nucleotide diversity (π) and diversity in haplotypes homozygosity tracts (H12) for Hap A (C) and Hap B (D) in japonica population. The vertical dotted black line shows the position of the TIP insertion. The OsGAP gene is schematically shown in blue. The horizontal dashed black line represents the mean of 1000 random permutation pulls (for p-value see Supplementary file 6).
Selection on TIP-eQTLs associated with OsMPH1 expression.
Representation of the two main OsMPH1 haplotypes present in rice and rufipogon/nivara populations identified in the rice super-pangenome. Conserved nucleotide regions are connected by gray marks. Structural variants longer than 50bp are shown as white spaces. TIP-eQTLs are shown as red boxes. B, Boxplot representation of the expression of the two OsMPH1 haplotypes in the japonica population. Numbers inside boxplots represent the number of accessions in each group. Analysis of nucleotide diversity (π) and diversity in haplotypes homozygosity tracts (H12) for Hap A in indica population (C) and in japonica population (D). The vertical dotted black line shows the position of the TIP insertion. The OsMPH1 gene, which is schematically shown in blue. The horizontal dashed black line represents the mean of 1000 random permutation pulls (for p-value see Supplementary file 6).
Expression variance explained by leading TIP-eQTL and SNP-eQTL in the associations detected for each gene (dot). Panels in A) represent the analysis performed using a combined genotype matrix including all SNPs and TIPs. Panels in B) represent the analysis performed using a combined genotype matrix including TIPs and a subsampled number of SNPs matching the exact number of TIPs.
Effect size of TIP-eQTLs based on their population frequency on the drought condition.
