Description of the pathogens included in the study, workflow overview and statistics of structure prediction.

A, Cladogram of the pathogens used in the study. The schematics represent the disease symptoms on their respective hosts, with the white areas representing galls. The secretome count indicates the number of proteins per species predicted to be secreted, and the functional annotation shows the percentage of the secretome predicted to contain a known protein domain in the Pfam, SUPERFAMILY, or Gene3D databases. B, Flowchart of the workflow used to predict the secretome and the corresponding 3D structures. C, Raincloud plot showing the median and density distribution of pLDDT scores of the predicted structures in each pathogen.

Visualization of dominant protein folds present in each pathogen.

A, Network plot of structurally similar secretome clusters with at least 15 members. Not all 255 clusters are shown to reduce complexity. Each node represents a single protein, and an edge between two nodes represents structural similarity (TMScore > 0.5). B, Representative structure of the dominant fold in each pathogen. Since Ankyrin repeats are common in both P. brassicae and S. subterranea, they are represented only once.

Sequence and structural similarity among HopQ1 homologs in U. maydis, A. candida and P. brassicae.

A, Network plot showing the structural similarity between the members of cluster 21. Edges denote structural similarity (TMScore > 0.5). B, Pairwise sequence identity between selected HopQ1 structural homologs from plasmodiophorids, oomycetes, and fungi, illustrating sequence dissimilarity between some proteins despite structural homology. C, Gene expression values (log2 TPM) of two highly induced P. brassicae genes at 16 and 26 dpi. D, 3D structure of the mature protein sequences, assuming a HopQ1-like fold.

Sequence and structural similarity among Mig1 homologs in Ustilago maydis.

A, Similarity matrix showing the pairwise sequence identity (%) between Mig1 cluster members. B, Similarity matrix showing the pairwise structural homology scores (TMScore) between Mig1 cluster members. C, Superimposition of two Mig1 homologues, illustrating structural similarity despite extreme sequence divergence. D, Differential gene expression patterns of two Mig1 tandem duplicates. E, Multiple alignment of protein sequences, highlighting the conservation of cysteine residues (marked in yellow). F, Visualization of the conserved cysteine residues forming disulfide bridges.

SUSS effector families are enriched in common motifs.

A, Network plots demonstrating that the two primary effector families in A. candida and S. endobioticum can only be grouped together when structural data is incorporated into the sequence-based clustering. The plots also indicate which sequence-based clusters contain the known effectors from these groups. B, ‘RAYH’ and ‘CCG’ motif patterns identified by MEME scan. C, Disulfide bridges in the ‘CCG’ motif, likely playing a pivotal role in structural maintenance, are highlighted in the virulence factor CCG30. A zoomed-in view of the ‘CCG module’ shows the four conserved cysteine residues forming disulfide bridge. D, The ‘RAYH’ motif, occupying the central position in the core alpha- helix bundle, is highlighted in six sequence-based subclusters within the AvrSen1-like cluster in S. endobioticum.

Diversity, structural features and host immune targets of ankyrin repeats in Plasmodiophorids.

A, Frequency of repeat-containing proteins in P. brassicae and S. subterranea. B, Network plot showing structural homology within P. brassicae Ankyrin repeats, also highlighting the Ankyrin repeats with SKP1/BTB/POZ superfamily domains. C, Alignment of Ankyrin motifs from P. brassicae and S. subterranea. D, Visualization of conserved hydrophobic residues in a single Ankyrin repeat module. E, Number of Ankyrin repeat proteins predicted to target Arabidopsis immune proteins. F, AlphaFold Multimer predicted complex of MPK3 and PBTT_00818, highlighting the predicted aligned errors of surface contacts under 4 Ångströms.