Hybrid genome assembly of the cannabis powdery mildew agentGolovinomyces ambrosiae uncovers important resources for deciphering virulence factors

Hybrid genome assembly of the cannabis powdery mildew agentGolovinomyces ambrosiae uncovers important resources for deciphering virulence factors | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Hybrid genome assembly of the cannabis powdery mildew agentGolovinomyces ambrosiae uncovers important resources for deciphering virulence factors Félix-Antoine Roy, Yanick Asselin, Caroline Labbé, Guillaume Quang Henri Nguyen, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9162455/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 9 You are reading this latest preprint version Abstract Background Cannabis powdery mildew, caused by the fungal pathogen Golovinomyces ambrosiae , poses a significant threat to licensed producers as its presence on marketable products can compromise product innocuity. While there is a focus in research for resistance genes within the Cannabis sativa germplasm, there is a lack of genetic information regarding the infecting agent, preventing validation of their effectiveness against different populations as the plant-pathogen interaction most likely follows a gene-for-gene relationship. In this paper, we assembled the first G. ambrosiae genome, providing insights into its genomic content and potential virulence determinants. The assembly was made using a hybrid approach, combining Oxford Nanopore Technologies long reads and Illumina short reads. Results The resulting 155.2 Mb genome is composed of 73 contigs, 13 scaffolds and has a completeness score of 97.5%. Subsequent analysis highlighted the substantial transposable elements content of the pathogen, occupying 82.64% of its genomic composition. Prediction of protein-coding genes revealed 6995 highly confident gene models, including 169 candidate effector proteins. Among the latter, we highlighted 14 candidates sharing key characteristics of confirmed effectors in other powdery mildew pathosystems such as the presence of signal peptides, RALPH-like domains, and Y/F/WxC motifs. Conclusions The result of this study provides valuable resources for future identification of avirulence genes within the G. ambrosiae species, responsible of conferring resistance when encoded effectors are recognized by a cognate host’s resistance genes. Those findings will lead to guided breeding strategies and, ultimately, the selection of cultivars adapted to the pathogen’s virulence profile. Powdery mildew Genome Cannabis sativa Genetic resistance Effector Avirulence Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Powdery mildews are part of the Ascomycota and are composed of an impressive diversity of species within the Erysiphales order [ 1 ]. To date, more than 900 species have been identified, and these obligate biotrophic fungi affect more than 16 000 angiosperms [ 2 ]. For cannabis ( Cannabis sativa L.), powdery mildew (PM) is one of the most common diseases in production facilities. In Canada only, production sites sampled from 2013 to 2017 revealed the presence of the infecting agent in 100% of locations surveyed [ 3 ]. The disease can lead to major yield consequences when infected leaves, stems and flowers reduce photosynthesis and nutrient uptake [4]. More importantly, due to microbial testing guidelines for the commercialisation of legal products, the presence of the pathogen on cannabis flowers can severely impact market access/regulatory approval [5]. There is a notable discrepancy across the literature regarding the nomenclature of the fungal pathogens causing powdery mildew on cannabis. Many reports have been referring to the fungus as Golovinomyces cichoracearum sensu lato ( Erysiphe cichoracearum ) [3, 6 ,7] while others have mentioned Golovinomyces spadiceus ( Erysiphe spadiceus ) [8, 9], Golovinomyces ambrosiae ( Erysiphe ambrosiae ) [10, 11] or Podosphaera macularis [12]. In the early 1900s, E. cichoracearum was considered to be a very cosmopolitan species, capable of infecting 280 different plant species divided into 27 families [13]. It was also noted that different biological forms were observed depending on the host. With this large host repertoire and the observation of different biological forms, it is likely that E. cichoracearum included many species presently classified under the genus Golovinomyces [ 14 ]. As of today, it is recognised that G. cichoracearum is a complex, now subdivided into 11 species including G. ambrosiae [ 2 ]. Furthermore, a recent multi-locus phylogenetic and taxonomic study revealed that despite G. ambrosiae and G. spadiceus being distinguished based on morphological traits [ 15 ], phylogenetic analysis using ITS, 28S rDNA, IGS, TUB2 and CHS1 sequences confirm that both represent a single taxon [ 14 ]. There is now a consensus that the only powdery mildew species affecting cannabis are G. ambrosiae and P. macularis , with the latter being documented only under field conditions. It has also recently been documented that both species can co-infect Cannabis sativa hosts [12, 16]. Many strategies have been adopted to prevent or reduce the disease pressure. These include climate control in indoor production facilities by creating unfavorable conditions for the growth of the fungus, such as proper ventilation, humidity adjustments, reduced plant density or the removal of infected plants/leaves [4, 17]. The use of organic fungicides can also be a solution for managing the disease, as several products are registered for this purpose in cannabis production [ 18 ]. Out of the 19 registered products in Canada, only a few have published data on their efficacy. While some organic fungicides show promising control of the disease, efficacy results appear to vary among trials [4, 18]. Moreover, successful management of powdery mildew with these products requires rigorous weekly applications, which are time-consuming and directly increase production costs. In many crops, powdery mildew is best managed by genetic approaches [4]. To our knowledge, two distinct mechanisms have been exploited to provide resistance to the disease [19]. One mechanism implicates resistance genes ( R genes ) with conserved nucleotide-binding site and/or leucine-rich repeat domains (NLRs), which encode proteins that function as intracellular immune receptors. These proteins are implicated in the direct or indirect recognition of specific secreted effector proteins encoded by the pathogen’s avirulence gene ( Avr ) and are responsible for triggering a hypersensitive reaction (HR), inducing localized cell death to prevent the spread of the infection. This model implicates a gene-for-gene interaction, as typically one R gene is effective against one specific Avr . On this basis, R genes provide race-specific powdery mildew immunity to the host and have been exploited in various crops such as wheat, pea, and grape [20, 21, 22]. The second mechanism is related to powdery mildew susceptibility ( S ) genes, referred to as Mildew resistance locus o (MLO). Unlike NLR genes, these are only found in plants, and they provide broad-spectrum resistance to powdery mildew pathogens when subjected to loss-of-function due to natural or induced mutations. This mechanism has been widely exploited for barley, since the discovery of a naturally mutated Mlo gene in an Ethiopian landrace in the 1930s, which provides complete powdery mildew resistance [19, 23]. Ultimately, both mechanisms provide an economically and environmentally friendly solution to prevent powdery mildew. Research for genetic resistance to powdery mildew in cannabis is currently advancing rapidly. In 2021, Mihalyov and Garfinkel were the first to identify a specific locus presumably containing an R gene, named PM1, in the Pacific Northwest experimental variety line “PNW39”, providing complete resistance to a powdery mildew isolate [ 24 ]. In the same year, Pépin et al. identified and characterized 15 MLO candidate genes across five different C. sativa genomes. Among those candidates, it was observed that two of them, CsMLO1 and CsMLO4, were significantly upregulated during the infection, making them strong candidates for susceptibility [ 25 ]. In 2024, Stack et al. discovered a major quantitative trait locus (QTL) neighboring the previously identified CsMLO1 on C. sativa chromosome 1 (Chr01) from the cultivar “FL 58”. By inspecting the CsMLO1 gene sequence on this cannabis variety, they discovered a 6.8 kb insertion that introduces a premature stop codon, reinforcing evidence that this candidate could contribute to powdery mildew resistance because of loss-of-function [26]. More recently, another research team discovered a second locus thought to harbour an R gene, designated PM2, through a bulk-segregant analysis and high-throughput RNA sequencing [27]. While these discoveries are promising for the development of elite powdery mildew resistant cannabis cultivars, the lack of genetic knowledge of the pathogen itself poses a significant hindrance for the validation of their efficacy in large scale deployment. Even though Mlo -based resistance is said to provide broad spectrum protection, it has been reported in other crops that some strains of the fungal pathogen can bypass this mechanism possibly due to intrinsic genetic divergences in a RBR-family E3 ubiquitin ligase and a medA-like transcriptional regulator. [ 28 ]. As for R genes, genetic knowledge of the pathogen is even more important given their intimate relationship with the pathogen’s Avr genes. Since Avrs are rapidly evolving genes [ 29 ], significant diversity among them is expected in different geographical locations or time periods, directly impacting the effectiveness of R genes. Ultimately, a deeper understanding of the genetic features of the pathogen will make it possible to proceed to functionality testing of the newly identified R genes to validate their performance in various pathological scenarios. In this study, we provide the first assembled genome of the cannabis powdery mildew pathogen Golovinomyces ambrosiae , produced by the combination of Oxford Nanopore Technologies (ONT) long reads and Illumina NovaSeq short reads sequencing platforms, providing insights into its genetic composition such as transposable elements content, protein coding genes and potential virulence factors. Material and Methods Collection of fungal material Heavily powdery mildew infected cannabis leaves were collected from the susceptible cultivar “Green dragon”, grown in a greenhouse compartment at Laval University, Québec, Canada. Whole leaves were selected, cut into leaflets and quickly dipped in a 5% acetate acetone solution (5g of cellulose acetate dissolved in 100 ml anhydrous acetone) [ 30 ]. Leaflets were dried for 8 min on two stainless steel cell spreaders placed side by side or until all acetone had completely evaporated leaving a hardened cellulose pellicle on the surface of the leaflets, trapping the mycelium and spores. The cellulose pellicles were peeled off the leaflet surfaces, stored in a 50-ml Falcon tube and placed immediately at -80°C until further use. DNA extraction A total of 500 mg of cellulose pellicles containing the fungal material were placed in a mortar with liquid nitrogen and crushed with a pestle until a fine powder was obtained. The improved High molecular weight (HMW) DNA extraction method developed by Russo et al. [ 31 , 32 ] was used for the current study. The final high-molecular-weight DNA concentration was quantified using the Qubit dsDNA BR Assay Kit on a Qubit Fluorometer (Thermo Fisher Scientific), and purity ratios were assessed with a NanoDrop spectrophotometer (Thermo Fisher Scientific). Fragment size distribution and DNA integrity were evaluated using the FemtoPulse system (Agilent Technologies) to confirm suitability for long-read sequencing. Library preparation and DNA sequencing For long-read sequencing, a library was prepared from 3 µg of HMW DNA using the ONT ligation sequencing kit (SQK-LSK114), following the manufacturer’s protocol, and enriched for fragments ≥ 3 kb using ONT’s Large Fragment Buffer. Sequencing was performed on a PromethION 2 solo (Oxford Nanopore Technologies, Oxford, UK) for 72 h using a FLO-PRO114M flow cell with R10.4.1 chemistry. From the same initial sample, 2 µg of DNA were sent to the Génome Québec Innovation Centre (Montréal, QC, Canada) where short-read library preparation was performed followed by a short-read Illumina NovaSeq sequencing to produce 2 x 150 paired-end reads. Hybrid genome assembly All steps used for genome assembly and subsequent analysis are summarized in Table 1 . Long reads were basecalled using ONT’s basecaller Dorado v1.0.2 within MinKNOW v25.04, with the basecalling model “ [email protected] ”. Adapter sequences were removed with Porechop v0.2.4 and reads were then filtered with NanoFilt v2.3.0 [33] to retain sequences with a Phred quality score equal or superior to 15 and a minimum length of 20 kb. To identify contaminants in the dataset, a sample of reads was queried against the BLASTn nt online database (E-value 1e-5). Species that could have been realistically present in the starting fungal sample, either by direct contact or present in the laboratory environment were selected for subsequent processing. Long and short reads were mapped using minimap2 v2.28 [34] to the selected contaminant genomes, ( C. sativa , Frankliniella occidentalis , Monilinia fructicola , Allantophomopsis cytisporea ) as well as genomes of closely related species of G. ambrosiae ( G. cichoracearum , Blumeria graminis f.sp. tritici and Erysiphe necator ) in order to capture reads closest to the target organism and eliminate unwanted sequences. Reads mapping only to powdery mildew genomes were used for the assembly using Flye v2.9.5 [35] (parameters: --nano-raw --iterations 3 --scaffold --min-overlap 8500). Illumina short reads were trimmed with Fastp 0.23.4 [36] and used for two rounds of polishing the long-read assembly using Pilon v1.23 [37]. To evaluate the completeness of the genome assembly, a Benchmarking Universal Single-copy Orthologs (BUSCO) [38] analysis was run using the Ascomycota lineage database (ascomycota_odb10). Phylogeny Golovinomyces ambrosiae was included in a genome-scale phylogenetic analysis using conserved single-copy orthologs identified with the BUSCO v5.8.2 pipeline and the ascomycota_odb10 lineage dataset. The BUSCO lineage comprises 3235 conserved orthologous groups. Genomic assemblies from representative taxa within Ascomycota were retrieved from the NCBI database. Accession number for each genome is provided in Additional file 1. BUSCO screening identified single-copy orthologs across all genomes. From the total ortholog set, 1805 genes detected in at least 90% of the analyzed genomes were retained for downstream phylogenomic reconstruction. For each retained ortholog, predicted protein sequences were aligned independently using MAFFT v7.525 with the parameter --auto. Individual alignments were subsequently concatenated into a supermatrix using AMAS v1.0[ 39 ] with the parameters concat -f fasta -d aa, resulting in a final concatenated amino acid (aa) matrix of 19.32M sites for 18 taxa, including G. ambrosiae and the selected Leotiomycetes representatives, as well as an appropriate outgroup. Maximum-likelihood phylogenetic inference was performed with IQ-TREE2 v2.1.3 [ 40 ]. Branch support was assessed using 1,000 ultrafast bootstrap replicates and 1,000 SH-aLRT tests. The best-fitting amino acid substitution mIQodel was selected automatically using ModelFinderPlus (-m MFP), based on the Bayesian Information Criterion (BIC). According to BIC, Q.plant + F+R7 was chosen as best-fit model. The phylogenetic tree was visualized using Interactive Tree of Life (ITOL) online platform [41]. Repetitive elements and protein-coding gene predictions The resulting genome was processed through RepeatModeler v2.0.6 [42] to identify repetitive regions which were then masked from the assembly using RepeatMasker v4.1.8 [43] to exclude those sequences from the gene prediction processes. Protein-coding genes were predicted using MAKER2 v2.31.10 [44], with provided evidence of gene annotations from the wheat powdery mildew agent Blumeria graminis f.sp. tritici and protein sequences of two closely related species of G. ambrosiae with genomic data available E. necator and G. cichoracearum . Ab initio gene predictions with Augustus v3.3.2 [45] were also integrated to the pipeline for additional gene models. The completeness of the gene prediction dataset was evaluated with BUSCO v5.8.2 against the Leotiomycetes database (leotiomycetes_odb10). Ultimately, a BLASTp search against the whole nonredundant protein database was ran to identify homologous protein sequences. Identification of candidate effector proteins (CEPs) Predicted protein sequences generated from the gene prediction pipeline were first filtered to keep only those ranging from 12 to 700 amino acids. The filtered dataset was submitted to the online SignalP 5.0 [46] servers to identify protein sequences with a predicted signal peptide. Sequences with the presence of a signal peptide were then processed by DeepTMHMM v2.0 [ 47 ] algorithm to predict transmembrane domains [ 48 ]. All protein sequences with a transmembrane domain were discarded and the remaining ones were submitted to the EffectorP 3.0 [ 49 ] online platform to ultimately predict effector-related proteins. To determine if any predicted sequences had homologs to known haustorium secreted proteins, a homology analysis was run using MMseqs2 [50] with publicly available RNA-seq data of Golovinomyces cichoracearum isolates GcM1 and GcM3 [51]. Final gene prediction filtering and annotation To obtain a final predicted gene dataset, all sequences validated by BUSCO analysis and all identified CEPs were initially kept. For the rest of the predicted genes, every sequence with an Annotation Edit Distance (AED) score superior to 0.1 was discarded to ensure high accuracy with evidence-based predictions. Functional annotation of this final gene set was runn using eggNOG-mapper v2.1.12 [52], InterProScan [53]. Additional search for specific domains was made using HHpred [54] online servers and HMMER from the EMBL-EBI website. Identification and classification of duplicated genes A BLASTp vs all (E-value < 1E-5) search was performed on the final protein sequence dataset of predicted genes with a maximum of five hits per sequence to identify duplicated genes. The BLASTp output was then filtered to keep only hits with a minimum identity of 80% and a minimal subject coverage of 75% to retain sequences with relatively low sequence divergence. The classification of these duplicated genes as dispersed, proximal or tandem was done using MCScanX [55] with the duplicate_gene_classifier script. Table 1 . Summary of tools used for the assembly and characterization of the Golovinomyces ambrosiae genome Workflow steps Tools Long read basecalling Dorado v1.0.2 – MinKNOW v25.04 Adapter sequences removal (long reads) Porechop v0.2.4 Adapter sequences removal (short reads) Fastp v0.23.4 Long reads quality and size filtering Nanofilt v2.3.0 Contaminant screening BLASTn nt online database Contaminant filtering Minimap2 v2.28 Long reads genome assembly Flye v2.9.5 Genome polishing Pilon v1.23 Genome completeness validation BUSCO v5.8.1 Phylogeny BUSCO v5.8.1, MAFFT v7.525, AMAS v.1.0, IQ-TREE v2.1.3 Repetitive regions identification RepeatModeler v2.0.6 Repetitive regions masking RepeatMasker v4.1.8 Protein coding gene prediction MAKER v2.31.10, AUGUSTUS v3.3.2 Protein coding gene annotation eggNOG-mapper v2.2.12, InterProScan, HHpred, HMMHER CEPs identification SignalP v5.0, DeepTMHMM v2.0, EffectorP 3.0 Duplicated genes identification BLASTp, MCScanX Results DNA sequencing and genome assembly The PromethION 2 sequencing process yielded a total of 46.85M reads, representing 99.42 Gb with an approximate N50 of 3.32 kb. After filtering for the desired Phred score, read length and the removal of sequences from other species, 163 035 reads were used for the long-read assembly. For the Illumina paired-end sequencing, a total of 70.3M read pairs were obtained, representing a total of 21.5 Gb of data. After all assembly processes, the final G. ambrosiae genome sequence was obtained with 33x coverage and a total size of 155.2 Mb, consisting of 73 contigs and 13 scaffolds. The assembly N50 is 3.5 Mb, and the largest sequence assembled is 12.0 Mb long. The BUSCO completeness analysis revealed that the assembly contains 97.5% (n = 1665) (96.4% single, n = 1645; 1,1% duplicated, n = 19) of the expected orthologs in the Ascomycota lineage (ascomycota_odb10), with a low proportion of fragmented (0.1%, n = 1) and missing genes (2.4%, n = 41). A summary of all assembly statistics is shown in Table 2 . Phylogeny The phylogenetic relationship of Golovinomyces ambrosiae with other Ascomycetes is displayed in Fig. 1 . Every node of the tree scored 100% bootstrap support, confirming excellent robustness for the grouping of each species. The species G. ambrosiae forms a monophyletic clade with other Golovinomyces species, and according to branch lengths, the tree shows that the newly assembled species is slightly less divergent from G. magnicellulatus compared to G. cichoracearum . This result supports the claim that G. ambrosiae is genetically distinct from G. cichoracearum . All species within the Golovinomyces genus share a most common ancestor with Erysiphe spp., which also form a monophyletic clade. Compared to all other powdery mildew species sampled, Golovinomyces spp. are most divergent from the grass powdery mildews Blumeria graminis f. sp. tritici and Blumeria graminis f. sp. hordei . As for the dicotyledonous powdery mildew pathogens only, species within the Podospharea genus are the most distant from G. ambrosiae. Transposable Elements Maximum-likelihood phylogenetic tree of 18 Ascomycetes including Golovinomyces ambrosiae . The phylogenetic tree illustrates the evolutionary relationship of G. ambrosiae with 17 other Ascomycetes. Species’ accession numbers are displayed in Additional file 1. The genome of every species was processed with BUSCO v5.8.1 using the ascomycota_10odb database. A total of 1805 genes, present in at least 90% of all genomes were retrieved and aligned using MAFFT v7.525. Alignments were concatenated by AMAS 1.0 to generate a supermatrix. The resulting amino acid supermatrix, consisting of 19.32M sites for all taxa was used for the generation of the maximum-likelihood phylogenetic tree using IQ-TREE v2.1.3. Grey labeled species are part of the outgroup used to root the phylogenetic tree, composed of three Ascomycetes ( Rhynchosporium graminicola , Phlyctema vagabunda and Botrytis fabae ) that are not part of the Leotiomycetes. All remaining species are related to the powdery mildew disease. Bootstrap support is shown in red below branches. Branches lengths, representing the number of amino acid substitutions per site are displayed over branches. The G. ambrosiae genome was largely composed of transposable elements (TE), with 82.64% repetitive regions. Retroelements dominate the TE content (57.62%), with Long Terminal Repeat (LTR) elements being the major class (29.34%), followed by Long Interspersed Nuclear Elements (LINEs) (28.28%). Among LTRs, Ty1/Copia and Gypsy/DIRS1 are the only two families present in this class. DNA transposons, including Tc1-IS630-Pogo, MULE-MuDR and hobo-Activator families make up 11.17% of the repetitive regions. A very small fraction accounted for Rolling-circles (0.02%), low complexity (0.02%) and simple repeats (0.48%). The remaining repeat content remained unclassified (13.32%). The complete relative T.E content and non-repetitive DNA of G. ambrosiae is illustrated in Fig. 2 . Genome composition of Golovinomyces ambrosiae. Relative proportion of transposable elements (TE) and non-repetitive DNA within the Golovinomyces ambrosiae genome. Repetitive regions were predicted with RepeatModeler v2.0.6. Retroelements occupy more than half of the genome size (57.62%), divided almost equally between Long Interspersed Nuclear Elements (LINEs) and Long Terminal Repeat (LTR) elements. DNA transposons are composed of Tc1-IS630-Pogo, MULE-MuDR, hobo-Activator and a small fraction of unclassified elements. A significant proportion (13.32%) of repetitive elements remain unclassified. In total, TEs make up 82.64% of the genome size (155.2 Mb), representing 128.27 Mb. Identification of candidate effector proteins (CEPs) and homology analysis A total of 169 CEPs were identified in the G. ambrosiae genome (Additional File 2). Of those, 122 were predicted to be primarily cytoplasmic, while 29 are expected to be apoplastic. The remaining 18 candidates were classified as dual-localized, meaning they received a prediction for both localizations. The smallest sequence predicted was 29 amino acid long, while the longest had a length of 687. The average size of all predicted CEPs was 232 aa long, with a median of 194. Among all predicted CEPs, 60 harbored a Y/F/WxC motif in their protein sequence. The frequencies of the three possible combinations of the motif were 32% for the YxC, 65% for the FxC and 3% for the WxC. Analysis performed with the RNA-seq data of predicted effectors upregulated in haustoria by G. cichoracearum revealed 23 homologs among the G. ambrosiae CEPs, with sequence identity ranging from 23.6% to 82.0% (Additional files 3). Almost half of all those homologous predicted effectors were found on contig 11. To validate if any of the predicted effectors in the entire dataset could be associated with RNase-Like Proteins associated with Haustoria (RALPH), a search was conducted in the protein annotations and revealed 24 CEPs sharing similarities with ribonucleases. More than half of these predicted sequences were found to harbor the Y/F/WxC motif, and additionally, 14 of them (CEP11_33.146, CEP11_58.17, CEP11_59.30, CEP11_60.3, CEP11_61.149, CEP11_62.1, CEP11_63.133, CEP11_63.94, CEP11_64.126, CEP11_65.8, CEP119_0.92, CEP119_18.91, CEP145_1.67, CEP255_7.25) were also part of the homologs to haustoria secreted CEPs of G. cichoracearum . In all these CEPs, at least one Y/F/WxC motif is present in the predicted RNase domain location. Based on the HHpred outputs, the predicted RNase domain for these CEPs shares homology with ribonuclease T1 of Aspergillus oryzae. The protein architecture of those 14 candidates is illustrated in Fig. 3 . Protein architecture of RALPH-like Candidate for Effector Proteins (CEPs) of Golovinomyces ambrosiae . Among predicted candidate effector proteins (CEPs), 14 shared the presence of a signal peptide, an RNase domain homologous to GUANYL-SPECIFIC RIBONUCLEASE T1 (E-value > 1e-5) and at least one Y/F/WxC motif. Three CEPs (CEP11_63.94, CEP11_64.126, CEP11_65.8) show identical protein sequences. Every CEP shown is also homologous to candidate effectors secreted by haustoria of Golovinomyces cichoracearum isolates GcM1 and GcM3. Signal peptide position was predicted by SignalP 5.0. RNase domains were predicted using HHpred online servers (E-value > 1e-5). Predicted protein coding genes and annotation Based on the prediction analysis, G. ambrosiae harbors a total of 6995 protein coding genes. Among those, 2767 are part of the 3234 core Leotimycetes genes (leotiomycetes_odb10). The functional annotation pipeline successfully characterized a total of 6021 predicted protein sequences by having an assigned eggNOG description (5477), GO terms (4692), KEGG pathways (1755) and/or Pfam domains (4710). Furthermore, InterProScan matched 5702 of the predicted protein sequences to known protein families, domains or conserved sites. The eggNOG-mapper tool additionally classified 3302 predicted protein sequences in Clusters of Orthologous Genes (COGs) functional categories. The category with the most abundant number of genes was Replication, recombination and repair, of which 76.72% were predicted to be implicated in transposable elements activity. This high number of genes in this category is consistent with the abundance of transposable elements composing the G. ambrosiae genome. Other categories with substantial number of genes are related to information storage and processing, cellular processes and signaling, and global metabolism. The 10 COG categories exhibiting the highest gene numbers are shown in Fig. 4 . Overall, a total of 974 predicted genes (13.92%) remained unclassified. The BLAST search against the nonredundant protein database identified 6748 (95.38%) of the total predicted protein sequences as having homologs within powdery mildew species, with the main ones being G. cichoracearum , G. magnicellulatus and Podosphaera aphanis . All protein annotation results are displayed in Additional file 4. Clusters of Orthologous Groups (COGs) functional categories with the highest gene numbers in Golovinomyces ambrosiae The protein annotation performed by eggNOG-mapper v2.2.12 classified 3302 genes in various COG categories. Predicted protein-coding genes grouped in the Replication, recombination and repair category are the most represented among all groups. Other groups with the highest gene counts are related to information storage and processing [J, A, K], cellular processes and signaling [T, U, O] and global metabolism [G, E, I]. Duplicated genes A total of 1780 genes were found to be duplicated within the G. ambrosiae genome, representing 25.45% of the whole predicted gene set. Most of those are dispersed (90.51%, n = 1611) while a fraction is categorised as proximal (7.13%, n = 127) or tandem duplicated (2.36%, n = 42). Based on COGs classification, many of those duplicated genes are part of the Replication, recombination and repair category. A manual inspection of the duplicated sequences in this category revealed 283 copies of a predicted gene, encoding a 716 to 790 amino acids protein with a mean identity of 99.66% (Additional file 5). Protein annotations from HMMER indicate that they all harbor a zinc-finger SWIM and a MULE related domain. Only a small proportion of duplicates (4.55%, n = 81) are associated with cellular processes and signaling or metabolism. In total, 18.09% of the identified duplicates remain unclassified, having no match with a corresponding eggNOG description, GO term, KEGG pathway, Pfam domain or InterProScan ID. Among CEPs, the proportions of duplicate classes were different from other protein coding genes. A total of 12.43% (n = 21) of CEPs showed duplication, with 52.38% (n = 11) being classified as proximal and the remaining 47.62% (n = 10) as dispersed. For proximal duplicates, there is a cluster on contig 11 of three CEPs (CEP11_63.94, CEP11_64.126, CEP11_65.8) with 100% sequence similarity. Still on the same contig, another CEP cluster showed four copies (CEP11_58.17, CEP11_33.146, CEP11_60.3, CEP11_63.133), of which two were proximal, and two dispersed. The predicted protein sequences of these four copies are 205 to 233 aa long and vary in similarity between 87.0% to 97.4%, possibly indicating a more ancient duplication. All these candidates are part of the RAPLH-like proteins, harbor the Y/F/WxC motif and are homologous to haustoria secreted candidate effector proteins of G. cichoracearum isolates GcM1 and GcM3. Other proximal duplications appear on contig 31, with four CEPs of near equal length and sequence similarity ranging from 80.25–95.12%. Table 2 Assembly statistics of the Golovinomyces ambrosiae genome Assembly metric Value Genome size (bp) 155 192 055 Number of total fragments 87 Number of contigs 73 Number of scaffolds 13 Assembly N50 (bp) 3 548 631 Longest fragment (bp) 12 043 756 Repeat content (%) 82.64 GC content (%) 41.57 Protein coding genes 6995 CEPs 169 Discussion The assembled genome presented in this study is among the most complete PM genomes available and provides the first genomic insights into the cannabis PM pathogen. Golovinomyces ambrosiae displays a very similar genome composition to other characterized powdery mildew species. Its size of 155.2 Mb is comparable to the 152.7 Mb of Podosphaera xanthii YZU573 [ 56 ], 129.9 Mb of G. magnicellulatus [57] and the predicted 173.8-221.8 Mb of G. cichoracearum [51]. The number of genes in the assembly is also consistent with other published PM genomes, ranging mostly from 6046 to 9372 [58, 59, 60]. While it seems common for these genomes to be dominated by transposable elements, and specifically by LINEs and LTRs, G. ambrosiae was found to possess the highest repetitive elements content among all published dicot PM genomes to date. With no evidence from the annotations of crucial genes required for repeat-induced point mutation (RIP), a mechanism widely employed among fungi for the disabling of TE [58, 61], their absence clearly supports the impressive invasion of repetitive regions in this cannabis pathogen. The abundance of TEs and the high number of genes implicated in their activity show remarkable genome plasticity in G. ambrosiae . In general, TEs can have negative impacts on genetic activities due to their potential of replicating in functionally essential DNA regions; on the other hand, they can also be drivers of beneficial adaptations and novel genes [62, 63, 64]. Their expansion can be accentuated by a lack of sexual recombination, which is typical of powdery mildew fungi that rely on asexual reproduction, a dominant condition in the context of indoor cannabis production. Pathogenicity may also be affected by the activity of TEs. Indeed, effector coding genes present in repeat-rich regions are more susceptible to accumulating mutations, which often improve fitness of the pathogen [62]. The detection of the high copy number of the MULE transposase and zinc SWIM finger domain-containing genes further demonstrates how dynamic the genome of this fungus may be. These characteristics are reminiscent of certain structural features related to the mudrA gene of maize, which encode the MURA protein, a putative transposase responsible of the transposition of the MULE-MuDR DNA transposons [65]. MULE-MuDRs are class II DNA transposons and have been shown in maize to be among the most mutagenic elements of all TEs [65]. Following a cut-and-paste transposition mechanism, it has been observed that they are not randomly inserted within the genome. In fact, when transposition occurs, MULE-MuDR have been found to have an insertion preference near the 5’ ends of genes and with recombinationally active regions of the genome. Further analyses of those genes would be interesting to validate if they are truly active and indeed have a role in transposition. In consideration of these observations and the abundance of TEs composing the assembled genome, it is reasonable to suggest that there is likely substantial genetic diversity within the G. ambrosiae species, potentially leading to isolates with differential virulence profiles. The genetic data regarding potential effector proteins provided by the assembly is also of great interest. Their abundance is in range with other identified CEPs of dicot PMs, particularly with the 159–175 found in G. cichoracearum [51]. Additionally, many candidate effectors of G. ambrosiae harbor the Y/F/WxC motif, which is a characteristic shared among powdery mildew effectors [66]. This motif has also been found to be an important structural element for RALPH proteins [67]. For CEPs involved in pathogenicity, the focus is on proteins that are secreted by the haustoria during pathogen-host interactions. The presence of a signal peptide, the Y/F/WxC motif and protein homology with haustoria secreted candidate effectors are important features that facilitate the filtration of the most promising candidates based on this characteristic. Additionally, since all known PM Avrs recognized by a corresponding host’s R gene are linked to RALPH proteins [58], CEPs sharing sequence similarity with ribonucleases is another significant characteristic worthy of consideration. With identified sequences sharing either one or more of these key attributes, this study may provide a strong foundation for future identification of bona fide Avr genes. Among those, the 14 CEPs sharing every of the Avr -like features mentioned as depicted in Fig. 3 appear to be the most promising. These candidates should be carefully studied in future transcriptomic analyses to first confirm their expression. Functional experiments should follow to confirm their role in pathogenicity. Of these CEPs, three were identified as proximal duplicates sharing perfect sequence identity. If these predicted candidates were found to be bona fide effectors implicated in virulence, this particular strain of G. ambrosiae may have great fitness advantages by benefiting from an increased effector production [ 68 ]. In contrast, it could be a significant drawback if those sequences are related to avirulent phenotypes, requiring gain-of-virulence mutations in every copy to counteract resistance [ 69 ]. The other CEPs showing four non-identical copies are also interesting. If implicated in virulence, this could be a sign that these genes were subject to diversifying selection, potentially as a result of an arms race against the host [70]. Direct Avr -R protein interactions have been validated in the Blumeria graminis f. sp hordei ( Bgh ) – barley pathosystem [71, 72, 73]. In total, seven effector proteins of Bgh are recognized by their matching Mildew Locus a (MLA) gene, all part of the Mla locus, encoding multiple allelic NLRs. In all studies, the selection of top-ranking candidate effectors has been made through transcriptome-wide association studies (TWAS) and/or comparative genome analysis, with the goal of finding significant sequence polymorphisms matching the distinct phenotypes among numerous Bgh isolates. For all effectors, the avirulent and virulent genotypes were either differentiated by nonsynonymous SNPs, presence/absence of transcript, a transposon insertion in the gene sequence or splice site mutations leading to intron retention. The process of matching Avr genes to their corresponding resistance protein in the host has also been performed in the wheat-powdery mildew Blumeria graminis f. sp tritici pathosystem. In the host’s genome, the Pm3 gene is the most diverse resistance gene studied, with 17 functional allelic variants identified to date. To those, three matching Avr genes have been identified, and gain-of-virulence mutations have been attributed to the presence of a suppressor of avirulence gene ( Svr ) and/or amino-acid polymorphisms [20, 74]. Other Avrs associated with Pm1a , Pm2 , Pm17 , Pm8 and Pm60 were also identified through QTL-mapping, genetic fine mapping, map-based cloning, genome-wide association studies, and avirulence depletion assay [69, 75, 76, 77, 78, 79]. In one of those studies, another gain-of-virulence mechanism such as complete deletion of the Avr gene has been highlighted [76]. These types of studies and their findings are critical for the comprehension of plant-pathogen interactions regarding virulence. Genetic characterisation of how resistance can hold or break down within certain pathogen populations is necessary for enlightened breeding strategies. In the context of cannabis powdery mildew, it is essential to extend the genomic profiling of multiple G. ambrosiae strains, which will allow researchers to better understand the diversity of genetic determinants implicated in virulence. Future work on the identification and cloning of the G. ambrosiae Avrs will allow functional validation against the currently identified R genes in the C. sativa germplasm. Subsequently, the unravelling of the distinct pathotypes will make it possible to confidently make use of the genetic resistance sources against powdery mildew. For wheat, it has been proposed to collectively create an interactive online R -gene atlas, which would provide information on exploited R gene and the corresponding Avr genotypes against which they provide protection [80]. Additionally, population survey data would be included to track changes in the effectorome of pathogens from different locations, informing on potential resistance breakdown threats. Since research on the cannabis powdery mildew resistance is still in its infancy, it would be of great interest to generate such data for each new resistance source discovered before findings get scattered. At least, for each newly discovered R gene in the host, equivalent efforts should be invested into deciphering the cognate Avr gene of the pathogen. In this context, this study sets a strong foundation for such future characterisations of host-pathogen interactions, enabling the most efficient approach of counteracting the disease. Abbreviations aa: Amino acid Avr: Avirulence gene Bgh: Blumeria graminis f. sp. hordei BIC: Bayesian Information Criterion BUSCO: Benchmarking Universal Single-copy Orthologs CEPs: Candidate effector proteins Chr01: Chromosome 1 COGs: Clusters of Orthologous genes HMW: High molecular weight HR: Hypersensitive reaction ITOL: Interactive Tree of Life MLA: Mildew Locus a MLO: Mildew resistance locus o ONT: Oxford Nanopore Technologies PM : Powdery mildew R genes : Resistance genes NLRs: Nucleotide-binding site and/or leucine-rich repeat domains QTL: Quantitative trait locus RALPH: RNase-Like Proteins associated with Haustoria RIP: Repeat-induced point mutation S: Susceptibility Svr: Suppressor of avirulence TWAS: Transcription-wide association studies Declarations Ethics, approval and consent to participate Not applicable Consent for publication Not applicable Availability of data and materials The genome assembly, Illumina paired-end reads, and Oxford Nanopore PromethION 2 reads have been deposited in the NCBI databases under BioProject accession number PRJNA1438116. The assembly is accessible at DDBJ/ENA/GenBank under the accession number JBWDLV000000000. The Illumina paired-end reads and Oxford Nanopore PromethION 2 have been deposited in NCBI Sequence Read Archive (SRA) under accession numbers SRR37663630 and SRR37669284. Data will be released publicly upon publication of the article. Reviewer link: https://dataview.ncbi.nlm.nih.gov/object/PRJNA1438116?reviewer=odqq7spltb4cjitk7vqmdq1dql Competing Interests The authors declare no competing interests Funding This research was supported by the Greenhouse Research Chair in Plant Protection-MAPAQ-Premier Tech and the program NSERC-Alliance awarded to RRB. RCL is funded by Genome Canada, Genome Quebec, CIHR and NFRF. FAR was supported by an NSERC Scholarship. Authors’ contributions F.A.R. performed all bioinformatics workflows, analyses and wrote the initial draft and manuscript. Y.A. and B.C. provided guidance for the bioinformatics workflows and reviewed the manuscript. C.L. and F.A.R. performed the DNA extraction. R.R.B. wrote, reviewed and edited the manuscript. D.T. edited the manuscript. Funding acquisition and supervision was provided by R.R.B. and D.T. R.C.L. devised the Nanopore strategies in genome analysis and revised the manuscript. S.M., J.G. and G.Q.H.N. revised the manuscript, used the PromethION Solo 2 for WGS and bioinformatics analyses. Acknowledgements Not applicable References Belanger, R. R., Bushnell, W. R., Dik, A. J., & Carver, T. L. (2002). The Powdery Mildews: a Comprehensive Treatise . American Phytopathological Society Press. Park, J. H., Choi, Y.-J., & Shin, H.-D. (2025). Revisiting Golovinomyces Species (Erysiphaceae) in Korea: Re-identification, New Records, and Description of Golovinomyces physalidis sp. nov. Mycobiology , 53 (4), 450–465. https://doi-org.acces.bibl.ulaval.ca/10.1080/12298093.2025.2517424 Pépin, N., Punja, Z. K., & Joly, D. L. (2018). Occurrence of Powdery Mildew Caused by Golovinomyces cichoracearum sensu lato on Cannabis sativa in Canada. Plant Disease, 102(12), 2644. https://doi-org.acces.bibl.ulaval.ca/10.1094/PDIS-04-18-0586-PDN Scott, & Punja. (2021). Evaluation of disease management approaches for powdery mildew on Cannabis sativa L. (marijuana) plants. Canadian Journal of Plant Pathology , 43 (3), 394–412. https://doi-org.acces.bibl.ulaval.ca/10.1080/07060661.2020.1836026 Health Canada (2025). Good production practices guide for cannabis . Government of Canada. http://epe.lac-bac.gc.ca/100/201/301/weekly_acquisitions_list-ef/2019/19-30/publications.gc.ca/collections/collection_2019/sc-hc/H14-271-2018-fra.pdf Maymon, M., Jerushalmi, S., & Freeman, S. (2020). First report of Golovinomyces cichoracearum sensu lato on Cannabis sativa in Israel. New Disease Reports , 42 (1), 11. https://doi-org.acces.bibl.ulaval.ca/10.5197/j.2044-0588.2020.042.011 Punja, Z. K., Collyer, D., Scott, C., Lung, S., Holmes, J., & Sutton, D. (2019). Pathogens and Molds Affecting Production and Quality of Cannabis sativa L. Frontiers in Plant Science , 10 , 1120. https://doi-org.acces.bibl.ulaval.ca/10.3389/fpls.2019.01120 Szarka, D., Tymon, L., Amsden, B., Dixon, E., Judy, J., & Gauthier, N. (2019). First Report of Powdery Mildew Caused by Golovinomyces spadiceus on Industrial Hemp ( Cannabis sativa) in Kentucky. Plant Disease , 103 (7), 1773. https://doi-org.acces.bibl.ulaval.ca/10.1094/PDIS-01-19-0049-PDN Farinas, C., & Hand, F. P. (2020). First Report of Golovinomyces spadiceus Causing Powdery Mildew on Industrial Hemp (Cannabis sativa) in Ohio. Plant Disease , 104 (10), 2727. Brochu, A.-S., Labbe, C., Bélanger, R. R., & Pérez-López, E. (2022). First Report of Powdery Mildew Caused by Golovinomyces ambrosiae on Cannabis sativa L. (marijuana) in Quebec, Canada. Plant Disease. https://doi-org.acces.bibl.ulaval.ca/10.1094/PDIS-02-22-0350-PDN Wiseman, M. S., Bates, T. A., Garfinkel, A. R., Ocamb, C. M., & Gent, D. H. (2021). First Report of Powdery Mildew Caused by Golovinomyces ambrosiae on Cannabis sativa in Oregon. Plant Disease , 105 (9), 2733. Punja, Z. K. (2022). First report of the powdery mildew pathogen of hops, Podosphaeria macularis, naturally infecting cannabis (Cannabis sativa L., marijuana) plants under field conditions. Canadian Journal of Plant Pathology , 44 (2), 235–249. https://doi-org.acces.bibl.ulaval.ca/10.1080/07060661.2021.1960424 Salmon, E. S. (1900). A MONOGRAPH OF THE ERYSIPHACEAE. Memoirs of the Torrey Botanical Club , 9 , 1–292. Qiu, P.-L., Liu, S.-Y., Bradshaw, M., Rooney-Latham, S., Takamatsu, S., Bulgakov, T. S., Tang, S.-R., Feng, J., Jin, D.-N., Aroge, T., Li, Y., Wang, L.-L., & Braun, U. (2020). Multi-locus phylogeny and taxonomy of an unresolved, heterogeneous species complex within the genus Golovinomyces (Ascomycota, Erysiphales), including G. ambrosiae, G. circumfusus and G. spadiceus. BMC Microbiology , 20 (1). https://doi-org.acces.bibl.ulaval.ca/10.1186/s12866-020-01731-9 Braun, U., & Cook, R. T. A. (2012). Taxonomic manual of the Erysiphales (powdery mildews) . CBS-KNAW Fungal Biodiversity Centre. Mieslerová, B., Kitner, M., Šenkyřík, J. B., Křivánková, T., Sedlářová, M., Braun, U., & Lebeda, A. (2024). The co-occurrence of two powdery mildew species on Cannabis sativa in the Czech Republic. Journal of Phytopathology, 172(3). https://doi-org.acces.bibl.ulaval.ca/10.1111/jph.13348 Buirs, L., & Punja, Z. K. (2024). Integrated Management of Pathogens and Microbes in Cannabis sativa L. (Cannabis) under Greenhouse Conditions. Plants , 13 (6), 786. https://doi.org/10.3390/plants13060786 Rufus J Arkinrinlola, Hansen. (2023). Efficacy of Organic Fungicides Against Hemp Powdery Mildew Caused by Golovinomyces ambrosiae in a Greenhouse in Tennessee. Sirangelo, T. M. (2023). NLR- and mlo-Based Resistance Mechanisms against Powdery Mildew in Cannabis sativa. Plants (Basel, Switzerland), 13(1). https://doi-org.acces.bibl.ulaval.ca/10.3390/plants13010105 Bourras, S., Kunz, L., Xue, M., Praz, C. R., Müller, M. C., Kälin, C., Schläfli, M., Ackermann, P., Flückiger, S., Parlange, F., Menardo, F., Schaefer, L. K., Ben-David, R., Roffler, S., Oberhaensli, S., Widrig, V., Lindner, S., Isaksson, J., Wicker, T., et al. (2019). The AvrPm3-Pm3 effector-NLR interactions control both race-specific resistance and host-specificity of cereal mildews on wheat. Nature Communications, 10(1), 2292. https://doi-org.acces.bibl.ulaval.ca/10.1038/s41467-019-10274-1 Sun, S., Fu, H., Wang, Z., Duan, C., Zong, X., Zhu, Z., & Singh, J. (2016). Discovery of a Novel er1 Allele Conferring Powdery Mildew Resistance in Chinese Pea (Pisum sativum L.) Landraces. Plos One , 11 (1). https://doi-org.acces.bibl.ulaval.ca/10.1371/journal.pone.0147624 Massonnet, M., Riaz, S., Pap, D., Figueroa-Balderas, R., Andrew Walker, M. & Cantu, D. (2022). The grape powdery mildew resistance loci Ren2, Ren3, Ren4D, Ren4U, Run1, Run1.2b, Run2.1, and Run2.2 activate different transcriptional responses to Erysiphe necator. Frontiers in Plant Science, 13. https://doi-org.acces.bibl.ulaval.ca/10.3389/fpls.2022.1096862 Jørgensen, I. H. (1992). Discovery, characterization and exploitation of Mlo powdery mildew resistance in barley. Euphytica : International Journal of Plant Breeding, 63(1-2), 141–152. https://doi-org.acces.bibl.ulaval.ca/10.1007/BF00023919 Mihalyov, P. D., & Garfinkel, A. R. (2021). Discovery and Genetic Mapping of PM1, a Powdery Mildew Resistance Gene in Cannabis sativa L. Frontiers in Agronomy, 3. https://doi-org.acces.bibl.ulaval.ca/10.3389/fagro.2021.720215 Pépin, N., Hebert, F. O., & Joly, D. L. (2021). Genome-Wide Characterization of the MLO Gene Family in Cannabis sativa Reveals Two Genes as Strong Candidates for Powdery Mildew Susceptibility. Frontiers in Plant Science , 12 , 729261. https://doi-org.acces.bibl.ulaval.ca/10.3389/fpls.2021.729261 Stack, G. M., Cala, A. R., Quade, M. A., Toth, J. A., Monserrate, L. A., Wilkerson, D. G., Carlson, C. H., Mamerto, A., Michael, T. P., Crawford, S., Smart, C. D., & Smart, L. B. (2024). Genetic Mapping, Identification, and Characterization of a Candidate Susceptibility Gene for Powdery Mildew in Cannabis sativa. Molecular Plant-Microbe Interactions : MPMI , 37 (1), 51–61. https://doi.org/10.1094/MPMI-04-23-0043-R Seifi, S., Leckie, K. M., Giles, I., O’Brien, T., MacKenzie, J. O., Todesco, M., Rieseberg, L. H., Baute, G. J., & Celedon, J. M. (2025). Mapping and characterization of a novel powdery mildew resistance locus (PM2) in Cannabis sativa L. Frontiers in Plant Science, 16, 1543229. https://doi-org.acces.bibl.ulaval.ca/10.3389/fpls.2025.1543229 Kusch, S., Frantzeskakis, L., Lassen, B. D., Kümmel, F., Pesch, L., Barsoum, M., Walden, K. D., & Panstruga, R. (2024). A fungal plant pathogen overcomes mlo-mediated broad-spectrum disease resistance by rapid gene loss. The New Phytologist, 244(3), 962–979. https://doi-org.acces.bibl.ulaval.ca/10.1111/nph.20063 Lo Presti, L., Lanver, D., Schweizer, G., Tanaka, S., Liang, L., Tollot, M., Zuccaro, A., Reissmann, S., & Kahmann, R. (2015). Fungal Effectors and Plant Susceptibility. Annual Review of Plant Biology, 66, 513–545. https://doi-org.acces.bibl.ulaval.ca/10.1146/annurev-arplant-043014-114623 Linhan Li, Benjamin Collier, & Pietro Spanu. (2019). Isolation of Powdery Mildew Haustoria from Infected Barley. Bio-Protocol, 9(14). https://doi-org.acces.bibl.ulaval.ca/10.21769/BioProtoc.3299 Mayjonade, B., Gouzy, J., Donnadieu, C., Pouilly, N., Marande, W., Callot, C., Langlade, N., & Muños, S. (2016). Extraction of high-molecular-weight genomic DNA for long-read sequencing of single molecules. BioTechniques, 61(4), 203–205. Russo, A., Mayjonade, B., Frei, D., Potente, G., Kellenberger, R. T., Frachon, L., Copetti, D., Studer, B., Frey, J. E., Grossniklaus, U., & Schlüter, P. M. (2022). Low-Input High-Molecular-Weight DNA Extraction for Long-Read Sequencing From Plants of Diverse Families. Frontiers in Plant Science, 13, 883897. https://doi-org.acces.bibl.ulaval.ca/10.3389/fpls.2022.883897 De Coster, W., D'Hert, S., Schultz, D. T., Cruts, M., & Van Broeckhoven, C. (2018). NanoPack: Visualizing and processing long-read sequencing data. Bioinformatics, 34(15), 2666–2669. https://doi.org/10.1093/bioinformatics/bty149 Li, H. (2018). Minimap2: Pairwise alignment for nucleotide sequences. Bioinformatics, 34(18), 3094–3100. https://doi.org/10.1093/bioinformatics/bty191 Kolmogorov, M., Yuan, J., Lin, Y., & Pevzner, P. A. (2019). Assembly of long, error-prone reads using repeat graphs. Nature Biotechnology, 37 (5), 540–546. https://doi.org/10.1038/s41587-019-0072-8 Chen, S. (2023). Ultrafast one-pass FASTQ data preprocessing, quality control, and deduplication using fastp. iMeta, 2 (2), e107. https://doi.org/10.1002/imt2.107 Walker, B. J., Abeel, T., Shea, T., Priest, M., Abouelliel, A., Sakthikumar, S., Cuomo, C. A., Zeng, Q., Wortman, J., Young, S. K., & Earl, A. M. (2014). Pilon: An integrated tool for comprehensive microbial variant detection and genome assembly improvement. PLoS ONE, 9 (11), e112963. https://doi.org/10.1371/journal.pone.0112963 Simão, F. A., Waterhouse, R. M., Ioannidis, P., Kriventseva, E. V., & Zdobnov, E. M. (2015). BUSCO: Assessing genome assembly and annotation completeness with single-copy orthologs. Bioinformatics, 31 (19), 3210–3212. https://doi.org/10.1093/bioinformatics/btv351 Borowiec M. L. (2016). AMAS: a fast tool for alignment manipulation and computing of summary statistics. PeerJ , 4 , e1660. https://doi-org.acces.bibl.ulaval.ca/10.7717/peerj.1660 Nguyen, L. T., Schmidt, H. A., von Haeseler, A., & Minh, B. Q. (2015). IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Molecular biology and evolution , 32 (1), 268–274. https://doi-org.acces.bibl.ulaval.ca/10.1093/molbev/msu300 Letunic and Bork (2024) Nucleic Acids Res doi: 10.1093/nar/gkae268 Flynn, J. M., Hubley, R., Goubert, C., Rosen, J., Clark, A. G., Feschotte, C., & Smit, A. F. (2020). RepeatModeler2 for automated genomic discovery of transposable element families. Proceedings of the National Academy of Sciences of the United States of America, 117(17), 9451–9457. https://doi.org/10.1073/pnas.1921046117 Chen, N. (2004). Using RepeatMasker to identify repetitive elements in genomic sequences. Current Protocols in Bioinformatics, 4 , Unit 4.10. Holt, C., & Yandell, M. (2011). MAKER2: an annotation pipeline and genome-database management tool for second-generation genome projects. BMC bioinformatics , 12 , 491. https://doi-org.acces.bibl.ulaval.ca/10.1186/1471-2105-12-491 Keller, O., Kollmar, M., Stanke, M., & Waack, S. (2011). A novel hybrid gene prediction method employing protein multiple sequence alignments. Bioinformatics (Oxford, England) , 27 (6), 757–763. https://doi-org.acces.bibl.ulaval.ca/10.1093/bioinformatics/btr010 Almagro Armenteros, J.J., Tsirigos, K.D., Sønderby, C.K. et al. SignalP 5.0 improves signal peptide predictions using deep neural networks. Nat Biotechnol 37 , 420–423 (2019). https://doi-org.acces.bibl.ulaval.ca/10.1038/s41587-019-0036-z Hallgren, J., Tsirigos, K. D., Pedersen, M. D., Almagro Armenteros, J. J., Marcatili, P., Nielsen, H., Krogh, A., & Winther, O. (2022). DeepTMHMM predicts alpha and beta transmembrane proteins using deep neural networks. bioRxiv . https://doi.org/10.1101/2022.04.08.487609 Sonah, H., Deshmukh, R. K., & Bélanger, R. R. (2016). Computational Prediction of Effector Proteins in Fungi: Opportunities and Challenges. Frontiers in Plant Science , 7 , 126. https://doi.org/10.3389/fpls.2016.00126 Sperschneider J and Dodds PN (2021) EffectorP 3.0: prediction of apoplastic and cytoplasmic effectors in fungi and oomycetes. MPMI . Steinegger, M., & Söding, J. (2017). MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nature Biotechnology, 35 (11), 1026–1028. Wu, Y., Ma, X., Pan, Z., Kale, S. D., Song, Y., King, H., Zhang, Q., Presley, C., Deng, X., Wei, C. I., & Xiao, S. (2018). Comparative genome analyses reveal sequence features reflecting distinct modes of host-adaptation between dicot and monocot powdery mildew. BMC genomics , 19 (1), 705. https://doi-org.acces.bibl.ulaval.ca/10.1186/s12864-018-5069-z Cantalapiedra, C. P., Hernández-Plaza, A., Letunic, I., Bork, P., & Huerta-Cepas, J. (2021). eggNOG-mapper v2: Functional Annotation, Orthology Assignments, and Domain Prediction at the Metagenomic Scale. Molecular biology and evolution , 38 (12), 5825–5829. https://doi-org.acces.bibl.ulaval.ca/10.1093/molbev/msab293 Jones P, Binns D, Chang H-Y, Fraser M, Li W, McAnulla C, McWilliam H, Maslen J, Mitchell A, Nuka G, et al. InterProScan 5: genome-scale protein function classification. Bioinformatics. 2014:30(9):1236–1240. 10.1093/bioinformatics/btu03. Zimmermann, L., Stephens, A., Nam, S. Z., Rau, D., Kübler, J., Lozajic, M., Gabler, F., Söding, J., Lupas, A. N., & Alva, V. (2018). A completely reimplemented MPI bioinformatics toolkit with a new HHpred server at its core. Journal of Molecular Biology, 430 (15), 2237–2243. https://doi.org/10.1016/j.jmb.2017.12.007 Wang, Y., Tang, H., DeBarry, J. D., Tan, X., Li, J., Wang, X., Lee, T., Jin, H., Marler, B., Guo, H., Kissinger, J. C., & Paterson, A. H. (2012). MCScanX: A toolkit for detection and evolutionary analysis of gene synteny and collinearity. Nucleic Acids Research, 40 (7), e49. https://doi.org/10.1093/nar/gkr1293 Wang, Z., Du, Y., Li, S., Xu, X., & Chen, X. (2023). A Complete Genome Sequence of Podosphaera xanthii Isolate YZU573, the Causal Agent of Powdery Mildew Isolated from Cucumber in China. Pathogens (Basel, Switzerland) , 12 (4), 561. https://doi-org.acces.bibl.ulaval.ca/10.3390/pathogens12040561 Farinas, C., Gluck-Thaler, E., Slot, J. C., & Peduto Hand, F. (2019). Whole-Genome Sequence of the Phlox Powdery Mildew Pathogen Golovinomyces magnicellulatus Strain FPH2017-1. Microbiology resource announcements , 8 (36), e00852-19. https://doi-org.acces.bibl.ulaval.ca/10.1128/MRA.00852-19 Kusch, S., Qian, J., Loos, A., Kümmel, F., Spanu, P. D., & Panstruga, R. (2024). Long-term and rapid evolution in powdery mildew fungi. Molecular ecology , 33 (10), e16909. https://doi-org.acces.bibl.ulaval.ca/10.1111/mec.16909 Frantzeskakis, L., Németh, M. Z., Barsoum, M., Kusch, S., Kiss, L., Takamatsu, S., & Panstruga, R. (2019). The Parauncinula polyspora Draft Genome Provides Insights into Patterns of Gene Erosion and Genome Expansion in Powdery Mildew Fungi. mBio , 10 (5), e01692-19. https://doi-org.acces.bibl.ulaval.ca/10.1128/mBio.01692-19 Gañán, L., White, R. A., 3rd, Friesen, M. L., Peever, T. L., & Amiri, A. (2020). A Genome Resource for the Apple Powdery Mildew Pathogen Podosphaera leucotricha . Phytopathology , 110 (11), 1756–1758. https://doi-org.acces.bibl.ulaval.ca/10.1094/PHYTO-05-20-0158-A Spanu, P. D. et al. (2010). Genome expansion and gene loss in powdery mildew fungi reveal tradeoffs in extreme parasitism. Science (New York, N.Y.) , 330 (6010), 1543–1546. https://doi-org.acces.bibl.ulaval.ca/10.1126/science.1194573 Möller, M., & Stukenbrock, E. H. (2017). Evolution and genome architecture in fungal plant pathogens. Nature reviews. Microbiology , 15 (12), 756–771. https://doi-org.acces.bibl.ulaval.ca/10.1038/nrmicro.2017.76 Nottensteiner, M., Zechmann, B., McCollum, C., & Hückelhoven, R. (2018). A barley powdery mildew fungus non-autonomous retrotransposon encodes a peptide that supports penetration success on barley. Journal of experimental botany , 69 (15), 3745–3758. https://doi-org.acces.bibl.ulaval.ca/10.1093/jxb/ery174 Sabelleck, B., & Panstruga, R. (2018). Novel jack-in-the-box effector of the barley powdery mildew pathogen?. Journal of experimental botany , 69 (15), 3511–3514. https://doi-org.acces.bibl.ulaval.ca/10.1093/jxb/ery192 Lisch, D. (2014). Mutator and MULE transposons. Microbiology Spectrum, 3(2), Article MDNA3-0032-2014. https://doi.org/10.1128/microbiolspec.MDNA3-0032-2014 Godfrey, D., Böhlenius, H., Pedersen, C., Zhang, Z., Emmersen, J., & Thordal-Christensen, H. (2010). Powdery mildew fungal effector candidates share N-terminal Y/F/WxC-motif. BMC genomics, 11, 317. https://doi-org.acces.bibl.ulaval.ca/10.1186/1471-2164-11-317 Cao, Y., Kümmel, F., Logemann, E., Gebauer, J. M., Lawson, A. W., Yu, D., Uthoff, M., Keller, B., Jirschitzka, J., Baumann, U., Tsuda, K., Chai, J., & Schulze-Lefert, P. (2023). Structural polymorphisms within a common powdery mildew effector scaffold as a driver of coevolution with cereal immune receptors. Proceedings of the National Academy of Sciences of the United States of America, 120(32), e2307604120. https://doi-org.acces.bibl.ulaval.ca/10.1073/pnas.2307604120 Pedersen, C., Ver Loren van Themaat, E., McGuffin, L. J., Abbott, J. C., Burgis, T. A., Barton, G., Bindschedler, L. V., Lu, X., Maekawa, T., Wessling, R., Cramer, R., Thordal-Christensen, H., Panstruga, R., & Spanu, P. D. (2012). Structure and evolution of barley powdery mildew effector candidates. BMC genomics , 13 , 694. https://doi-org.acces.bibl.ulaval.ca/10.1186/1471-2164-13-694 Müller, M. C., Kunz, L., Schudel, S., Lawson, A. W., Kammerecker, S., Isaksson, J., Wyler, M., Graf, J., Sotiropoulos, A. G., Praz, C. R., Manser, B., Wicker, T., Bourras, S., & Keller, B. (2022). Ancient variation of the AvrPm17 gene in powdery mildew limits the effectiveness of the introgressed rye Pm17 resistance gene in wheat. Proceedings of the National Academy of Sciences of the United States of America, 119 (30), Article e2108808119. https://doi.org/10.1073/pnas.2108808119 Liu, L., Xu, L., Jia, Q., Pan, R., Oelmüller, R., Zhang, W., & Wu, C. (2019). Arms race: diverse effector proteins with conserved motifs. Plant signaling & behavior , 14 (2), 1557008. https://doi-org.acces.bibl.ulaval.ca/10.1080/15592324.2018.1557008 Lu, X., Kracher, B., Saur, I. M. L., Bauer, S., Ellwood, S. R., Wise, R., Yaeno, T., Maekawa, T., & Schulze-Lefert, P. (2016). Allelic barley MLA immune receptors recognize sequence-unrelated avirulence effectors of the powdery mildew pathogen. Proceedings of the National Academy of Sciences of the United States of America, 113 (36), E6486–E6495. https://doi.org/10.1073/pnas.1612947113 Saur, I. M. L., Bauer, S., Kracher, B., Lu, X., Franzeskakis, L., Müller, M. C., Sabelleck, B., Kümmel, F., & Panstruga, R. (2019). Multiple pairs of allelic MLA immune receptor-powdery mildew AVRA effectors argue for a direct recognition mechanism. eLife, 8 , Article e44471. https://doi.org/10.7554/eLife.44471 Bauer, S., Yu, D., Lawson, A. W., Saur, I. M. L., Frantzeskakis, L., Kracher, B., Logemann, E., Chai, J., Maekawa, T., & Schulze-Lefert, P. (2021). The leucine-rich repeats in allelic barley MLA immune receptors define specificity towards sequence-unrelated powdery mildew avirulence effectors with a predicted common RNase-like fold. PLoS Pathogens, 17 (1), Article e1009223. https://doi.org/10.1371/journal.ppat.1009223 Bourras, S., McNally, K. E., Ben-David, R., Parlange, F., Roffler, S., Praz, C. R., Oberhaensli, S., Menardo, F., Stirnweis, D., Frenkel, Z., Schaefer, L. K., Flückiger, S., Treier, G., Herren, G., Korol, A. B., Wicker, T., & Keller, B. (2015). Multiple avirulence loci and allele-specific effector recognition control the Pm3 race-specific resistance of wheat to powdery mildew. The Plant Cell, 27 (10), 2991–3012. https://doi.org/10.1105/tpc.15.00171 Hewitt, T., Müller, M.C., Molnár, I., Mascher, M., Holušová, K., Šimková, H., Kunz, L., Zhang, J., Li, J., Bhatt, D., Sharma, R., Schudel, S., Yu, G., Steuernagel, B., Periyannan, S., Wulff, B., Ayliffe, M., McIntosh, R., Keller, B., Lagudah, E. and Zhang, P. (2021), A highly differentiated region of wheat chromosome 7AL encodes a Pm1a immune receptor that recognizes its corresponding AvrPm1a effector from Blumeria graminis . New Phytol, 229: 2812-2826. https://doi-org.acces.bibl.ulaval.ca/10.1111/nph.17075 Praz, C. R., Bourras, S., Zeng, F., Sánchez-Martín, J., Menardo, F., Xue, M., Yang, L., Roffler, S., Böni, R., Herren, G., McNally, K. E., Ben-David, R., Parlange, F., Oberhaensli, S., Flückiger, S., Schäfer, L. K., Wicker, T., Yu, D., & Keller, B. (2017). AvrPm2 encodes an RNase-like avirulence effector which is conserved in the two different specialized forms of wheat and rye powdery mildew fungus. New Phytologist, 213 , 1301–1314. https://doi.org/10.1111/nph.14372 Manser, B., Koller, T., Praz, C. R., Roulin, A. C., Zbinden, H., Arora, S., Steuernagel, B., Wulff, B. B. H., Keller, B., & Sánchez-Martín, J. (2021). Identification of specificity-defining amino acids of the wheat immune receptor Pm2 and powdery mildew effector AvrPm2. The Plant Journal, 106 (4), 993–1007. https://doi.org/10.1111/tpj.15214 Kunz, L., Sotiropoulos, A. G., Graf, J., Razavi, M., Keller, B., & Müller, M. C. (2023). The broad use of the Pm8 resistance gene in wheat resulted in hypermutation of the AvrPm8 gene in the powdery mildew pathogen. BMC biology , 21 (1), 29. https://doi-org.acces.bibl.ulaval.ca/10.1186/s12915-023-01513-5 Kunz, L., Jigisha, J., Menardo, F., Sotiropoulos, A. G., Zbinden, H., Zou, S., Tang, D., Hückelhoven, R., Keller, B., & Müller, M. C. (2025). Avirulence depletion assay: Combining R gene-mediated selection with bulk sequencing for rapid avirulence gene identification in wheat powdery mildew. PLoS pathogens , 21 (1), e1012799. https://doi-org.acces.bibl.ulaval.ca/10.1371/journal.ppat.1012799 Hafeez, A. N., Arora, S., Ghosh, S., Gilbert, D., Bowden, R. L., & Wulff, B. B. H. (2021). Creation and judicious application of a wheat resistance gene atlas. Molecular plant , 14 (7), 1053–1070. https://doi-org.acces.bibl.ulaval.ca/10.1016/j.molp.2021.05.014 Additional Declarations No competing interests reported. Supplementary Files AdditionalfilesGa.xlsx Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 22 Apr, 2026 Reviews received at journal 19 Apr, 2026 Reviews received at journal 17 Apr, 2026 Reviewers agreed at journal 11 Apr, 2026 Reviewers agreed at journal 10 Apr, 2026 Reviewers invited by journal 08 Apr, 2026 Editor assigned by journal 02 Apr, 2026 Submission checks completed at journal 31 Mar, 2026 First submitted to journal 31 Mar, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9162455","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":621394666,"identity":"3bef29d1-703e-45f6-8276-1c761dafb623","order_by":0,"name":"Félix-Antoine Roy","email":"","orcid":"","institution":"Université Laval","correspondingAuthor":false,"prefix":"","firstName":"Félix-Antoine","middleName":"","lastName":"Roy","suffix":""},{"id":621394674,"identity":"98e9012b-8b06-46cb-85f4-7d339e0029bc","order_by":1,"name":"Yanick Asselin","email":"","orcid":"","institution":"Université Laval","correspondingAuthor":false,"prefix":"","firstName":"Yanick","middleName":"","lastName":"Asselin","suffix":""},{"id":621394675,"identity":"25f762e2-3db3-4268-ad57-16f3a9575ec8","order_by":2,"name":"Caroline Labbé","email":"","orcid":"","institution":"Université Laval","correspondingAuthor":false,"prefix":"","firstName":"Caroline","middleName":"","lastName":"Labbé","suffix":""},{"id":621394678,"identity":"15c676f6-3705-498a-8dae-bfbb2cad88f3","order_by":3,"name":"Guillaume Quang Henri Nguyen","email":"","orcid":"","institution":"Université Laval","correspondingAuthor":false,"prefix":"","firstName":"Guillaume","middleName":"Quang Henri","lastName":"Nguyen","suffix":""},{"id":621394686,"identity":"93e6d1f1-44f2-4e06-ad47-3823d924fbce","order_by":4,"name":"Sima Mohammadi","email":"","orcid":"","institution":"Université Laval","correspondingAuthor":false,"prefix":"","firstName":"Sima","middleName":"","lastName":"Mohammadi","suffix":""},{"id":621394693,"identity":"c660a147-9662-4c9d-98e1-18aa1f12773a","order_by":5,"name":"Jeff Gauthier","email":"","orcid":"","institution":"Université Laval","correspondingAuthor":false,"prefix":"","firstName":"Jeff","middleName":"","lastName":"Gauthier","suffix":""},{"id":621394704,"identity":"1b0981ca-a3f0-4382-8c67-1b1f213efa4c","order_by":6,"name":"Benjamin Cinget","email":"","orcid":"","institution":"Université Laval","correspondingAuthor":false,"prefix":"","firstName":"Benjamin","middleName":"","lastName":"Cinget","suffix":""},{"id":621394708,"identity":"b4329d5a-ef40-4508-b317-668ec4413466","order_by":7,"name":"Davoud Torkamaneh","email":"","orcid":"","institution":"Université Laval","correspondingAuthor":false,"prefix":"","firstName":"Davoud","middleName":"","lastName":"Torkamaneh","suffix":""},{"id":621394710,"identity":"2f8ffc71-2620-47d8-8c6a-c1bca9205813","order_by":8,"name":"Roger C. Levesque","email":"","orcid":"","institution":"Université Laval","correspondingAuthor":false,"prefix":"","firstName":"Roger","middleName":"C.","lastName":"Levesque","suffix":""},{"id":621394712,"identity":"ec7d7881-248b-4ff4-85a7-fb8c96dd35b7","order_by":9,"name":"Richard R. Bélanger","email":"data:image/png;base64,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","orcid":"","institution":"Université Laval","correspondingAuthor":true,"prefix":"","firstName":"Richard","middleName":"R.","lastName":"Bélanger","suffix":""}],"badges":[],"createdAt":"2026-03-18 19:08:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9162455/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9162455/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107481006,"identity":"fc3762fc-1222-4956-ac3a-82abe4d74acb","added_by":"auto","created_at":"2026-04-22 02:15:05","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":41072,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMaximum-likelihood phylogenetic tree of 18 Ascomycetes including \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eGolovinomyces ambrosiae\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e.\u003c/strong\u003e \u003cbr\u003e\nThe phylogenetic tree illustrates the evolutionary relationship of \u003cem\u003eG. ambrosiae\u003c/em\u003e with 17 other Ascomycetes.\u003cbr\u003e\nSpecies’ accession numbers are displayed in Additional file 1. The genome of every species was processed with BUSCO v5.8.1 using the ascomycota_10odb database. A total of 1805 genes, present in at least 90% of all genomes were retrieved and aligned using MAFFT v7.525. Alignments were concatenated by AMAS 1.0 to generate a supermatrix. The resulting amino acid supermatrix, consisting of 19.32M sites for all taxa was used for the generation of the maximum-likelihood phylogenetic tree using IQ-TREE v2.1.3. \u0026nbsp;Grey labeled species are part of the outgroup used to root the phylogenetic tree, composed of three Ascomycetes (\u003cem\u003eRhynchosporium graminicola\u003c/em\u003e, \u003cem\u003ePhlyctema vagabunda and\u003c/em\u003e \u003cem\u003eBotrytis fabae\u003c/em\u003e) that are not part of the Leotiomycetes. All remaining species are related to the powdery mildew disease. Bootstrap support is shown in red below branches. Branches lengths, representing the number of amino acid substitutions per site are displayed over branches.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-9162455/v1/d9ed28746f75584385df3a98.png"},{"id":107482894,"identity":"f3aac491-f28a-4fe5-ab33-3ba86a79ae83","added_by":"auto","created_at":"2026-04-22 02:25:23","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":76451,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGenome composition of Golovinomyces ambrosiae. \u003cbr\u003e\n \u003c/strong\u003eRelative proportion of transposable elements (TE) and non-repetitive DNA within the Golovinomyces ambrosiae genome. Repetitive regions were predicted with RepeatModeler v2.0.6. Retroelements occupy more than half of the genome size (57.62%), divided almost equally between Long Interspersed Nuclear Elements (LINEs) and Long Terminal Repeat (LTR) elements. DNA transposons are composed of Tc1-IS630-Pogo, MULE-MuDR, hobo-Activator and a small fraction of unclassified elements. A significant proportion (13.32%) of repetitive elements remain unclassified. In total, TEs make up 82.64% of the genome size (155.2 Mb), representing 128.27 Mb.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-9162455/v1/3183cbb33865e4766ea7300c.png"},{"id":107095987,"identity":"57f9a525-8329-4c01-ac21-569d0d08d1d3","added_by":"auto","created_at":"2026-04-16 17:13:55","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":74822,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eProtein architecture of RALPH-like Candidate for Effector Proteins (CEPs) of Golovinomyces ambrosiae.\u003cbr\u003e\n \u003c/strong\u003eAmong predicted candidate effector proteins (CEPs), 14 shared the presence of a signal peptide, an RNase domain homologous to GUANYL-SPECIFIC RIBONUCLEASE T1 (E-value \u0026gt; 1e-5) and at least one Y/F/WxC motif. Three CEPs (CEP11_63.94, CEP11_64.126, CEP11_65.8) show identical protein sequences. Every CEP shown is also homologous to candidate effectors secreted by haustoria of Golovinomyces cichoracearum isolates GcM1 and GcM3. Signal peptide position was predicted by SignalP 5.0. RNase domains were predicted using HHpred online servers (E-value \u0026gt; 1e-5).\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-9162455/v1/d4545a2fce625ece26f8e656.png"},{"id":107481728,"identity":"64aacfc1-528b-4e2b-87d1-22ffd15c17db","added_by":"auto","created_at":"2026-04-22 02:19:49","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":89677,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eClusters of Orthologous Groups (COGs) functional categories with the highest gene numbers in Golovinomyces ambrosiae\u003cbr\u003e\n \u003c/strong\u003eThe protein annotation performed by eggNOG-mapper v2.2.12 classified 3302 genes in various COG categories. Predicted protein-coding genes grouped in the Replication, recombination and repair category are the most represented among all groups. Other groups with the highest gene counts are related to information storage and processing [J, A, K], cellular processes and signaling [T, U, O] and global metabolism [G, E, I].\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-9162455/v1/9b43a88c1441298b7ada2c7a.png"},{"id":109220268,"identity":"92d122e5-8865-4b5d-bc23-db3da097d84e","added_by":"auto","created_at":"2026-05-13 20:16:02","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":607905,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9162455/v1/92abdca7-b442-4195-91c2-0fcc58baa573.pdf"},{"id":107095984,"identity":"8a1ff5cb-7966-4769-b250-4debfd9adcc5","added_by":"auto","created_at":"2026-04-16 17:13:55","extension":"xlsx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":3493982,"visible":true,"origin":"","legend":"","description":"","filename":"AdditionalfilesGa.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9162455/v1/e237a10fbbec6200ca812986.xlsx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Hybrid genome assembly of the cannabis powdery mildew agentGolovinomyces ambrosiae uncovers important resources for deciphering virulence factors","fulltext":[{"header":"Introduction","content":"\u003cp\u003ePowdery mildews are part of the Ascomycota and are composed of an impressive diversity of species within the Erysiphales order [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. To date, more than 900 species have been identified, and these obligate biotrophic fungi affect more than 16 000 angiosperms [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. For cannabis (\u003cem\u003eCannabis sativa\u003c/em\u003e L.), powdery mildew (PM) is one of the most common diseases in production facilities. In Canada only, production sites sampled from 2013 to 2017 revealed the presence of the infecting agent in 100% of locations surveyed [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. The disease can lead to major yield consequences when infected leaves, stems and flowers reduce photosynthesis and nutrient uptake [4]. More importantly, due to microbial testing guidelines for the commercialisation of legal products, the presence of the pathogen on cannabis flowers can severely impact market access/regulatory approval [5].\u003c/p\u003e \u003cp\u003eThere is a notable discrepancy across the literature regarding the nomenclature of the fungal pathogens causing powdery mildew on cannabis. Many reports have been referring to the fungus as \u003cem\u003eGolovinomyces cichoracearum sensu lato\u003c/em\u003e (\u003cem\u003eErysiphe cichoracearum\u003c/em\u003e) [3, 6 ,7] while others have mentioned \u003cem\u003eGolovinomyces spadiceus\u003c/em\u003e (\u003cem\u003eErysiphe spadiceus\u003c/em\u003e) [8, 9], \u003cem\u003eGolovinomyces ambrosiae\u003c/em\u003e (\u003cem\u003eErysiphe ambrosiae\u003c/em\u003e) [10, 11] or \u003cem\u003ePodosphaera macularis\u003c/em\u003e [12]. In the early 1900s, \u003cem\u003eE. cichoracearum\u003c/em\u003e was considered to be a very cosmopolitan species, capable of infecting 280 different plant species divided into 27 families [13]. It was also noted that different biological forms were observed depending on the host. With this large host repertoire and the observation of different biological forms, it is likely that \u003cem\u003eE. cichoracearum\u003c/em\u003e included many species presently classified under the genus \u003cem\u003eGolovinomyces\u003c/em\u003e [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. As of today, it is recognised that \u003cem\u003eG. cichoracearum\u003c/em\u003e is a complex, now subdivided into 11 species including \u003cem\u003eG. ambrosiae\u003c/em\u003e [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Furthermore, a recent multi-locus phylogenetic and taxonomic study revealed that despite \u003cem\u003eG. ambrosiae\u003c/em\u003e and \u003cem\u003eG. spadiceus\u003c/em\u003e being distinguished based on morphological traits [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e15\u003c/span\u003e], phylogenetic analysis using ITS, 28S rDNA, IGS, TUB2 and CHS1 sequences confirm that both represent a single taxon [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. There is now a consensus that the only powdery mildew species affecting cannabis are \u003cem\u003eG. ambrosiae\u003c/em\u003e and \u003cem\u003eP. macularis\u003c/em\u003e, with the latter being documented only under field conditions. It has also recently been documented that both species can co-infect \u003cem\u003eCannabis sativa\u003c/em\u003e hosts [12, 16].\u003c/p\u003e \u003cp\u003eMany strategies have been adopted to prevent or reduce the disease pressure. These include climate control in indoor production facilities by creating unfavorable conditions for the growth of the fungus, such as proper ventilation, humidity adjustments, reduced plant density or the removal of infected plants/leaves [4, 17]. The use of organic fungicides can also be a solution for managing the disease, as several products are registered for this purpose in cannabis production [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Out of the 19 registered products in Canada, only a few have published data on their efficacy. While some organic fungicides show promising control of the disease, efficacy results appear to vary among trials [4, 18]. Moreover, successful management of powdery mildew with these products requires rigorous weekly applications, which are time-consuming and directly increase production costs.\u003c/p\u003e \u003cp\u003eIn many crops, powdery mildew is best managed by genetic approaches [4]. To our knowledge, two distinct mechanisms have been exploited to provide resistance to the disease [19]. One mechanism implicates resistance genes (\u003cem\u003eR genes\u003c/em\u003e) with conserved nucleotide-binding site and/or leucine-rich repeat domains (NLRs), which encode proteins that function as intracellular immune receptors. These proteins are implicated in the direct or indirect recognition of specific secreted effector proteins encoded by the pathogen\u0026rsquo;s avirulence gene (\u003cem\u003eAvr\u003c/em\u003e) and are responsible for triggering a hypersensitive reaction (HR), inducing localized cell death to prevent the spread of the infection. This model implicates a gene-for-gene interaction, as typically one \u003cem\u003eR\u003c/em\u003e gene is effective against one specific \u003cem\u003eAvr\u003c/em\u003e. On this basis, \u003cem\u003eR\u003c/em\u003e genes provide race-specific powdery mildew immunity to the host and have been exploited in various crops such as wheat, pea, and grape [20, 21, 22]. The second mechanism is related to powdery mildew susceptibility (\u003cem\u003eS\u003c/em\u003e) genes, referred to as \u003cem\u003eMildew resistance locus o\u003c/em\u003e (MLO). Unlike \u003cem\u003eNLR\u003c/em\u003e genes, these are only found in plants, and they provide broad-spectrum resistance to powdery mildew pathogens when subjected to loss-of-function due to natural or induced mutations. This mechanism has been widely exploited for barley, since the discovery of a naturally mutated \u003cem\u003eMlo\u003c/em\u003e gene in an Ethiopian landrace in the 1930s, which provides complete powdery mildew resistance [19, 23]. Ultimately, both mechanisms provide an economically and environmentally friendly solution to prevent powdery mildew.\u003c/p\u003e \u003cp\u003eResearch for genetic resistance to powdery mildew in cannabis is currently advancing rapidly. In 2021, Mihalyov and Garfinkel were the first to identify a specific locus presumably containing an \u003cem\u003eR\u003c/em\u003e gene, named PM1, in the Pacific Northwest experimental variety line \u0026ldquo;PNW39\u0026rdquo;, providing complete resistance to a powdery mildew isolate [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. In the same year, P\u0026eacute;pin et al. identified and characterized 15 MLO candidate genes across five different \u003cem\u003eC. sativa\u003c/em\u003e genomes. Among those candidates, it was observed that two of them, CsMLO1 and CsMLO4, were significantly upregulated during the infection, making them strong candidates for susceptibility [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. In 2024, Stack et al. discovered a major quantitative trait locus (QTL) neighboring the previously identified CsMLO1 on \u003cem\u003eC. sativa\u003c/em\u003e chromosome 1 (Chr01) from the cultivar \u0026ldquo;FL 58\u0026rdquo;. By inspecting the CsMLO1 gene sequence on this cannabis variety, they discovered a 6.8 kb insertion that introduces a premature stop codon, reinforcing evidence that this candidate could contribute to powdery mildew resistance because of loss-of-function [26]. More recently, another research team discovered a second locus thought to harbour an \u003cem\u003eR\u003c/em\u003e gene, designated PM2, through a bulk-segregant analysis and high-throughput RNA sequencing [27].\u003c/p\u003e \u003cp\u003eWhile these discoveries are promising for the development of elite powdery mildew resistant cannabis cultivars, the lack of genetic knowledge of the pathogen itself poses a significant hindrance for the validation of their efficacy in large scale deployment. Even though \u003cem\u003eMlo\u003c/em\u003e-based resistance is said to provide broad spectrum protection, it has been reported in other crops that some strains of the fungal pathogen can bypass this mechanism possibly due to intrinsic genetic divergences in a RBR-family E3 ubiquitin ligase and a medA-like transcriptional regulator. [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. As for \u003cem\u003eR\u003c/em\u003e genes, genetic knowledge of the pathogen is even more important given their intimate relationship with the pathogen\u0026rsquo;s \u003cem\u003eAvr\u003c/em\u003e genes. Since \u003cem\u003eAvrs\u003c/em\u003e are rapidly evolving genes [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e29\u003c/span\u003e], significant diversity among them is expected in different geographical locations or time periods, directly impacting the effectiveness of \u003cem\u003eR\u003c/em\u003e genes. Ultimately, a deeper understanding of the genetic features of the pathogen will make it possible to proceed to functionality testing of the newly identified \u003cem\u003eR\u003c/em\u003e genes to validate their performance in various pathological scenarios.\u003c/p\u003e \u003cp\u003eIn this study, we provide the first assembled genome of the cannabis powdery mildew pathogen \u003cem\u003eGolovinomyces ambrosiae\u003c/em\u003e, produced by the combination of Oxford Nanopore Technologies (ONT) long reads and Illumina NovaSeq short reads sequencing platforms, providing insights into its genetic composition such as transposable elements content, protein coding genes and potential virulence factors.\u003c/p\u003e"},{"header":"Material and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eCollection of fungal material\u003c/h2\u003e \u003cp\u003eHeavily powdery mildew infected cannabis leaves were collected from the susceptible cultivar \u0026ldquo;Green dragon\u0026rdquo;, grown in a greenhouse compartment at Laval University, Qu\u0026eacute;bec, Canada. Whole leaves were selected, cut into leaflets and quickly dipped in a 5% acetate acetone solution (5g of cellulose acetate dissolved in 100 ml anhydrous acetone) [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. Leaflets were dried for 8 min on two stainless steel cell spreaders placed side by side or until all acetone had completely evaporated leaving a hardened cellulose pellicle on the surface of the leaflets, trapping the mycelium and spores. The cellulose pellicles were peeled off the leaflet surfaces, stored in a 50-ml Falcon tube and placed immediately at -80\u0026deg;C until further use.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eDNA extraction\u003c/h3\u003e\n\u003cp\u003eA total of 500 mg of cellulose pellicles containing the fungal material were placed in a mortar with liquid nitrogen and crushed with a pestle until a fine powder was obtained. The improved High molecular weight (HMW) DNA extraction method developed by Russo et al. [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e32\u003c/span\u003e] was used for the current study. The final high-molecular-weight DNA concentration was quantified using the Qubit dsDNA BR Assay Kit on a Qubit Fluorometer (Thermo Fisher Scientific), and purity ratios were assessed with a NanoDrop spectrophotometer (Thermo Fisher Scientific). Fragment size distribution and DNA integrity were evaluated using the FemtoPulse system (Agilent Technologies) to confirm suitability for long-read sequencing.\u003c/p\u003e\n\u003ch3\u003eLibrary preparation and DNA sequencing\u003c/h3\u003e\n\u003cp\u003eFor long-read sequencing, a library was prepared from 3 \u0026micro;g of HMW DNA using the ONT ligation sequencing kit (SQK-LSK114), following the manufacturer\u0026rsquo;s protocol, and enriched for fragments\u0026thinsp;\u0026ge;\u0026thinsp;3 kb using ONT\u0026rsquo;s Large Fragment Buffer. Sequencing was performed on a PromethION 2 solo (Oxford Nanopore Technologies, Oxford, UK) for 72 h using a FLO-PRO114M flow cell with R10.4.1 chemistry. From the same initial sample, 2 \u0026micro;g of DNA were sent to the G\u0026eacute;nome Qu\u0026eacute;bec Innovation Centre (Montr\u0026eacute;al, QC, Canada) where short-read library preparation was performed followed by a short-read Illumina NovaSeq sequencing to produce 2 x 150 paired-end reads.\u003c/p\u003e\n\u003ch3\u003eHybrid genome assembly\u003c/h3\u003e\n\u003cp\u003eAll steps used for genome assembly and subsequent analysis are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Long reads were basecalled using ONT\u0026rsquo;s basecaller Dorado v1.0.2 within MinKNOW v25.04, with the basecalling model \u0026ldquo;[email protected]\u0026rdquo;. Adapter sequences were removed with Porechop v0.2.4 and reads were then filtered with NanoFilt v2.3.0 [33] to retain sequences with a Phred quality score equal or superior to 15 and a minimum length of 20 kb. To identify contaminants in the dataset, a sample of reads was queried against the BLASTn \u003cem\u003ent\u003c/em\u003e online database (E-value 1e-5). Species that could have been realistically present in the starting fungal sample, either by direct contact or present in the laboratory environment were selected for subsequent processing. Long and short reads were mapped using minimap2 v2.28 [34] to the selected contaminant genomes, (\u003cem\u003eC. sativa\u003c/em\u003e, \u003cem\u003eFrankliniella occidentalis\u003c/em\u003e, \u003cem\u003eMonilinia fructicola\u003c/em\u003e, \u003cem\u003eAllantophomopsis cytisporea\u003c/em\u003e) as well as genomes of closely related species of \u003cem\u003eG. ambrosiae\u003c/em\u003e (\u003cem\u003eG. cichoracearum\u003c/em\u003e, \u003cem\u003eBlumeria graminis\u003c/em\u003e f.sp. \u003cem\u003etritici\u003c/em\u003e and \u003cem\u003eErysiphe necator\u003c/em\u003e) in order to capture reads closest to the target organism and eliminate unwanted sequences. Reads mapping only to powdery mildew genomes were used for the assembly using Flye v2.9.5 [35] (parameters: --nano-raw --iterations 3 --scaffold --min-overlap 8500). Illumina short reads were trimmed with Fastp 0.23.4 [36] and used for two rounds of polishing the long-read assembly using Pilon v1.23 [37]. To evaluate the completeness of the genome assembly, a Benchmarking Universal Single-copy Orthologs (BUSCO) [38] analysis was run using the Ascomycota lineage database (ascomycota_odb10).\u003c/p\u003e\n\u003ch3\u003ePhylogeny\u003c/h3\u003e\n\u003cp\u003e \u003cem\u003eGolovinomyces ambrosiae\u003c/em\u003e was included in a genome-scale phylogenetic analysis using conserved single-copy orthologs identified with the BUSCO v5.8.2 pipeline and the ascomycota_odb10 lineage dataset. The BUSCO lineage comprises 3235 conserved orthologous groups. Genomic assemblies from representative taxa within Ascomycota were retrieved from the NCBI database. Accession number for each genome is provided in Additional file 1. BUSCO screening identified single-copy orthologs across all genomes. From the total ortholog set, 1805 genes detected in at least 90% of the analyzed genomes were retained for downstream phylogenomic reconstruction. For each retained ortholog, predicted protein sequences were aligned independently using MAFFT v7.525 with the parameter --auto. Individual alignments were subsequently concatenated into a supermatrix using AMAS v1.0[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e39\u003c/span\u003e] with the parameters concat -f fasta -d aa, resulting in a final concatenated amino acid (aa) matrix of 19.32M sites for 18 taxa, including \u003cem\u003eG. ambrosiae\u003c/em\u003e and the selected Leotiomycetes representatives, as well as an appropriate outgroup.\u003c/p\u003e \u003cp\u003eMaximum-likelihood phylogenetic inference was performed with IQ-TREE2 v2.1.3 [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. Branch support was assessed using 1,000 ultrafast bootstrap replicates and 1,000 SH-aLRT tests. The best-fitting amino acid substitution mIQodel was selected automatically using ModelFinderPlus (-m MFP), based on the Bayesian Information Criterion (BIC). According to BIC, Q.plant\u0026thinsp;+\u0026thinsp;F+R7 was chosen as best-fit model. The phylogenetic tree was visualized using Interactive Tree of Life (ITOL) online platform [41].\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eRepetitive elements and protein-coding gene predictions\u003c/h2\u003e \u003cp\u003eThe resulting genome was processed through RepeatModeler v2.0.6 [42] to identify repetitive regions which were then masked from the assembly using RepeatMasker v4.1.8 [43] to exclude those sequences from the gene prediction processes. Protein-coding genes were predicted using MAKER2 v2.31.10 [44], with provided evidence of gene annotations from the wheat powdery mildew agent \u003cem\u003eBlumeria graminis\u003c/em\u003e f.sp. \u003cem\u003etritici\u003c/em\u003e and protein sequences of two closely related species of \u003cem\u003eG. ambrosiae\u003c/em\u003e with genomic data available \u003cem\u003eE. necator\u003c/em\u003e and \u003cem\u003eG. cichoracearum\u003c/em\u003e. Ab initio gene predictions with Augustus v3.3.2 [45] were also integrated to the pipeline for additional gene models. The completeness of the gene prediction dataset was evaluated with BUSCO v5.8.2 against the Leotiomycetes database (leotiomycetes_odb10). Ultimately, a BLASTp search against the whole nonredundant protein database was ran to identify homologous protein sequences.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eIdentification of candidate effector proteins (CEPs)\u003c/h3\u003e\n\u003cp\u003ePredicted protein sequences generated from the gene prediction pipeline were first filtered to keep only those ranging from 12 to 700 amino acids. The filtered dataset was submitted to the online SignalP 5.0 [46] servers to identify protein sequences with a predicted signal peptide. Sequences with the presence of a signal peptide were then processed by DeepTMHMM v2.0 [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e47\u003c/span\u003e] algorithm to predict transmembrane domains [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e48\u003c/span\u003e]. All protein sequences with a transmembrane domain were discarded and the remaining ones were submitted to the EffectorP 3.0 [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e49\u003c/span\u003e] online platform to ultimately predict effector-related proteins. To determine if any predicted sequences had homologs to known haustorium secreted proteins, a homology analysis was run using MMseqs2 [50] with publicly available RNA-seq data of \u003cem\u003eGolovinomyces cichoracearum\u003c/em\u003e isolates GcM1 and GcM3 [51].\u003c/p\u003e\n\u003ch3\u003eFinal gene prediction filtering and annotation\u003c/h3\u003e\n\u003cp\u003eTo obtain a final predicted gene dataset, all sequences validated by BUSCO analysis and all identified CEPs were initially kept. For the rest of the predicted genes, every sequence with an Annotation Edit Distance (AED) score superior to 0.1 was discarded to ensure high accuracy with evidence-based predictions. Functional annotation of this final gene set was runn using eggNOG-mapper v2.1.12 [52], InterProScan [53]. Additional search for specific domains was made using HHpred [54] online servers and HMMER from the EMBL-EBI website.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eIdentification and classification of duplicated genes\u003c/h2\u003e \u003cp\u003eA BLASTp vs all (E-value\u0026thinsp;\u0026lt;\u0026thinsp;1E-5) search was performed on the final protein sequence dataset of predicted genes with a maximum of five hits per sequence to identify duplicated genes. The BLASTp output was then filtered to keep only hits with a minimum identity of 80% and a minimal subject coverage of 75% to retain sequences with relatively low sequence divergence. The classification of these duplicated genes as dispersed, proximal or tandem was done using MCScanX [55] with the \u003cem\u003eduplicate_gene_classifier\u003c/em\u003e script.\u003c/p\u003e \u003cp\u003e\u003cstrong\u003eTable\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003cstrong\u003e.\u0026nbsp;\u003cstrong\u003eSummary of tools used for the assembly and characterization of the\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eGolovinomyces ambrosiae\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;genome\u003c/strong\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"630\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eWorkflow steps\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 328px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTools\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003eLong read basecalling\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 328px;\"\u003e\n \u003cp\u003eDorado v1.0.2 \u0026ndash; MinKNOW v25.04\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003eAdapter sequences removal (long reads)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 328px;\"\u003e\n \u003cp\u003ePorechop v0.2.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003eAdapter sequences removal (short reads)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 328px;\"\u003e\n \u003cp\u003eFastp v0.23.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003eLong reads quality and size filtering\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 328px;\"\u003e\n \u003cp\u003eNanofilt v2.3.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003eContaminant screening\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 328px;\"\u003e\n \u003cp\u003eBLASTn nt online database\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003eContaminant filtering\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 328px;\"\u003e\n \u003cp\u003eMinimap2 v2.28\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003eLong reads genome assembly\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 328px;\"\u003e\n \u003cp\u003eFlye v2.9.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003eGenome polishing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 328px;\"\u003e\n \u003cp\u003ePilon v1.23\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003eGenome completeness validation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 328px;\"\u003e\n \u003cp\u003eBUSCO v5.8.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003ePhylogeny\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 328px;\"\u003e\n \u003cp\u003eBUSCO v5.8.1, MAFFT v7.525, AMAS v.1.0, IQ-TREE v2.1.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003eRepetitive regions identification\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 328px;\"\u003e\n \u003cp\u003eRepeatModeler v2.0.6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003eRepetitive regions masking\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 328px;\"\u003e\n \u003cp\u003eRepeatMasker v4.1.8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003eProtein coding gene prediction\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 328px;\"\u003e\n \u003cp\u003eMAKER v2.31.10, AUGUSTUS v3.3.2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003eProtein coding gene annotation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 328px;\"\u003e\n \u003cp\u003eeggNOG-mapper v2.2.12, InterProScan, HHpred, HMMHER\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003eCEPs identification\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 328px;\"\u003e\n \u003cp\u003eSignalP v5.0, DeepTMHMM v2.0, EffectorP 3.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003eDuplicated genes identification\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 328px;\"\u003e\n \u003cp\u003eBLASTp, MCScanX\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eDNA sequencing and genome assembly\u003c/h2\u003e \u003cp\u003eThe PromethION 2 sequencing process yielded a total of 46.85M reads, representing 99.42 Gb with an approximate N50 of 3.32 kb. After filtering for the desired Phred score, read length and the removal of sequences from other species, 163 035 reads were used for the long-read assembly. For the Illumina paired-end sequencing, a total of 70.3M read pairs were obtained, representing a total of 21.5 Gb of data. After all assembly processes, the final \u003cem\u003eG. ambrosiae\u003c/em\u003e genome sequence was obtained with 33x coverage and a total size of 155.2 Mb, consisting of 73 contigs and 13 scaffolds. The assembly N50 is 3.5 Mb, and the largest sequence assembled is 12.0 Mb long. The BUSCO completeness analysis revealed that the assembly contains 97.5% (n\u0026thinsp;=\u0026thinsp;1665) (96.4% single, n\u0026thinsp;=\u0026thinsp;1645; 1,1% duplicated, n\u0026thinsp;=\u0026thinsp;19) of the expected orthologs in the Ascomycota lineage (ascomycota_odb10), with a low proportion of fragmented (0.1%, n\u0026thinsp;=\u0026thinsp;1) and missing genes (2.4%, n\u0026thinsp;=\u0026thinsp;41). A summary of all assembly statistics is shown in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003ePhylogeny\u003c/h2\u003e \u003cp\u003eThe phylogenetic relationship of \u003cem\u003eGolovinomyces ambrosiae\u003c/em\u003e with other Ascomycetes is displayed in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Every node of the tree scored 100% bootstrap support, confirming excellent robustness for the grouping of each species. The species \u003cem\u003eG. ambrosiae\u003c/em\u003e forms a monophyletic clade with other \u003cem\u003eGolovinomyces\u003c/em\u003e species, and according to branch lengths, the tree shows that the newly assembled species is slightly less divergent from \u003cem\u003eG. magnicellulatus\u003c/em\u003e compared to \u003cem\u003eG. cichoracearum\u003c/em\u003e. This result supports the claim that \u003cem\u003eG. ambrosiae\u003c/em\u003e is genetically distinct from \u003cem\u003eG. cichoracearum\u003c/em\u003e. All species within the \u003cem\u003eGolovinomyces\u003c/em\u003e genus share a most common ancestor with \u003cem\u003eErysiphe\u003c/em\u003e spp., which also form a monophyletic clade. Compared to all other powdery mildew species sampled, \u003cem\u003eGolovinomyces\u003c/em\u003e spp. are most divergent from the grass powdery mildews \u003cem\u003eBlumeria graminis\u003c/em\u003e f. sp. \u003cem\u003etritici\u003c/em\u003e and \u003cem\u003eBlumeria graminis\u003c/em\u003e f. sp. \u003cem\u003ehordei\u003c/em\u003e. As for the dicotyledonous powdery mildew pathogens only, species within the \u003cem\u003ePodospharea\u003c/em\u003e genus are the most distant from \u003cem\u003eG. ambrosiae.\u003c/em\u003e\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eTransposable Elements\u003c/h2\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eMaximum-likelihood phylogenetic tree of 18 Ascomycetes including\u003c/b\u003e \u003cb\u003eGolovinomyces ambrosiae\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eThe phylogenetic tree illustrates the evolutionary relationship of \u003cem\u003eG. ambrosiae\u003c/em\u003e with 17 other Ascomycetes.\u003c/p\u003e \u003cp\u003eSpecies\u0026rsquo; accession numbers are displayed in Additional file 1. The genome of every species was processed with BUSCO v5.8.1 using the ascomycota_10odb database. A total of 1805 genes, present in at least 90% of all genomes were retrieved and aligned using MAFFT v7.525. Alignments were concatenated by AMAS 1.0 to generate a supermatrix. The resulting amino acid supermatrix, consisting of 19.32M sites for all taxa was used for the generation of the maximum-likelihood phylogenetic tree using IQ-TREE v2.1.3. Grey labeled species are part of the outgroup used to root the phylogenetic tree, composed of three Ascomycetes (\u003cem\u003eRhynchosporium graminicola\u003c/em\u003e, \u003cem\u003ePhlyctema vagabunda and Botrytis fabae\u003c/em\u003e) that are not part of the Leotiomycetes. All remaining species are related to the powdery mildew disease. Bootstrap support is shown in red below branches. Branches lengths, representing the number of amino acid substitutions per site are displayed over branches.\u003c/p\u003e \u003cp\u003eThe \u003cem\u003eG. ambrosiae\u003c/em\u003e genome was largely composed of transposable elements (TE), with 82.64% repetitive regions. Retroelements dominate the TE content (57.62%), with Long Terminal Repeat (LTR) elements being the major class (29.34%), followed by Long Interspersed Nuclear Elements (LINEs) (28.28%). Among LTRs, Ty1/Copia and Gypsy/DIRS1 are the only two families present in this class. DNA transposons, including Tc1-IS630-Pogo, MULE-MuDR and hobo-Activator families make up 11.17% of the repetitive regions. A very small fraction accounted for Rolling-circles (0.02%), low complexity (0.02%) and simple repeats (0.48%). The remaining repeat content remained unclassified (13.32%). The complete relative T.E content and non-repetitive DNA of \u003cem\u003eG. ambrosiae\u003c/em\u003e is illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eGenome composition of\u003c/b\u003e \u003cb\u003eGolovinomyces ambrosiae.\u003c/b\u003e\u003c/p\u003e \u003cp\u003eRelative proportion of transposable elements (TE) and non-repetitive DNA within the \u003cem\u003eGolovinomyces ambrosiae\u003c/em\u003e genome. Repetitive regions were predicted with RepeatModeler v2.0.6. Retroelements occupy more than half of the genome size (57.62%), divided almost equally between Long Interspersed Nuclear Elements (LINEs) and Long Terminal Repeat (LTR) elements. DNA transposons are composed of Tc1-IS630-Pogo, MULE-MuDR, hobo-Activator and a small fraction of unclassified elements. A significant proportion (13.32%) of repetitive elements remain unclassified. In total, TEs make up 82.64% of the genome size (155.2 Mb), representing 128.27 Mb.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eIdentification of candidate effector proteins (CEPs) and homology analysis\u003c/h2\u003e \u003cp\u003eA total of 169 CEPs were identified in the \u003cem\u003eG. ambrosiae\u003c/em\u003e genome (Additional File 2). Of those, 122 were predicted to be primarily cytoplasmic, while 29 are expected to be apoplastic. The remaining 18 candidates were classified as dual-localized, meaning they received a prediction for both localizations. The smallest sequence predicted was 29 amino acid long, while the longest had a length of 687. The average size of all predicted CEPs was 232 aa long, with a median of 194. Among all predicted CEPs, 60 harbored a Y/F/WxC motif in their protein sequence. The frequencies of the three possible combinations of the motif were 32% for the YxC, 65% for the FxC and 3% for the WxC. Analysis performed with the RNA-seq data of predicted effectors upregulated in haustoria by \u003cem\u003eG. cichoracearum\u003c/em\u003e revealed 23 homologs among the \u003cem\u003eG. ambrosiae\u003c/em\u003e CEPs, with sequence identity ranging from 23.6% to 82.0% (Additional files 3). Almost half of all those homologous predicted effectors were found on contig 11. To validate if any of the predicted effectors in the entire dataset could be associated with RNase-Like Proteins associated with Haustoria (RALPH), a search was conducted in the protein annotations and revealed 24 CEPs sharing similarities with ribonucleases. More than half of these predicted sequences were found to harbor the Y/F/WxC motif, and additionally, 14 of them (CEP11_33.146, CEP11_58.17, CEP11_59.30, CEP11_60.3, CEP11_61.149, CEP11_62.1, CEP11_63.133, CEP11_63.94, CEP11_64.126, CEP11_65.8, CEP119_0.92, CEP119_18.91, CEP145_1.67, CEP255_7.25) were also part of the homologs to haustoria secreted CEPs of \u003cem\u003eG. cichoracearum\u003c/em\u003e. In all these CEPs, at least one Y/F/WxC motif is present in the predicted RNase domain location. Based on the HHpred outputs, the predicted RNase domain for these CEPs shares homology with ribonuclease T1 of \u003cem\u003eAspergillus oryzae.\u003c/em\u003e The protein architecture of those 14 candidates is illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eProtein architecture of RALPH-like Candidate for Effector Proteins (CEPs) of\u003c/b\u003e \u003cb\u003eGolovinomyces ambrosiae\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eAmong predicted candidate effector proteins (CEPs), 14 shared the presence of a signal peptide, an RNase domain homologous to GUANYL-SPECIFIC RIBONUCLEASE T1 (E-value\u0026thinsp;\u0026gt;\u0026thinsp;1e-5) and at least one Y/F/WxC motif. Three CEPs (CEP11_63.94, CEP11_64.126, CEP11_65.8) show identical protein sequences. Every CEP shown is also homologous to candidate effectors secreted by haustoria of \u003cem\u003eGolovinomyces cichoracearum\u003c/em\u003e isolates GcM1 and GcM3. Signal peptide position was predicted by SignalP 5.0. RNase domains were predicted using HHpred online servers (E-value\u0026thinsp;\u0026gt;\u0026thinsp;1e-5).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003ePredicted protein coding genes and annotation\u003c/h2\u003e \u003cp\u003eBased on the prediction analysis, \u003cem\u003eG. ambrosiae\u003c/em\u003e harbors a total of 6995 protein coding genes. Among those, 2767 are part of the 3234 core Leotimycetes genes (leotiomycetes_odb10). The functional annotation pipeline successfully characterized a total of 6021 predicted protein sequences by having an assigned eggNOG description (5477), GO terms (4692), KEGG pathways (1755) and/or Pfam domains (4710). Furthermore, InterProScan matched 5702 of the predicted protein sequences to known protein families, domains or conserved sites. The eggNOG-mapper tool additionally classified 3302 predicted protein sequences in Clusters of Orthologous Genes (COGs) functional categories. The category with the most abundant number of genes was Replication, recombination and repair, of which 76.72% were predicted to be implicated in transposable elements activity. This high number of genes in this category is consistent with the abundance of transposable elements composing the \u003cem\u003eG. ambrosiae\u003c/em\u003e genome. Other categories with substantial number of genes are related to information storage and processing, cellular processes and signaling, and global metabolism. The 10 COG categories exhibiting the highest gene numbers are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. Overall, a total of 974 predicted genes (13.92%) remained unclassified. The BLAST search against the nonredundant protein database identified 6748 (95.38%) of the total predicted protein sequences as having homologs within powdery mildew species, with the main ones being \u003cem\u003eG. cichoracearum\u003c/em\u003e, \u003cem\u003eG. magnicellulatus\u003c/em\u003e and \u003cem\u003ePodosphaera aphanis\u003c/em\u003e. All protein annotation results are displayed in Additional file 4.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eClusters of Orthologous Groups (COGs) functional categories with the highest gene numbers in \u003cem\u003eGolovinomyces ambrosiae\u003c/em\u003e\u003c/h2\u003e \u003cp\u003eThe protein annotation performed by eggNOG-mapper v2.2.12 classified 3302 genes in various COG categories. Predicted protein-coding genes grouped in the Replication, recombination and repair category are the most represented among all groups. Other groups with the highest gene counts are related to information storage and processing [J, A, K], cellular processes and signaling [T, U, O] and global metabolism [G, E, I].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eDuplicated genes\u003c/h2\u003e \u003cp\u003eA total of 1780 genes were found to be duplicated within the \u003cem\u003eG. ambrosiae\u003c/em\u003e genome, representing 25.45% of the whole predicted gene set. Most of those are dispersed (90.51%, n\u0026thinsp;=\u0026thinsp;1611) while a fraction is categorised as proximal (7.13%, n\u0026thinsp;=\u0026thinsp;127) or tandem duplicated (2.36%, n\u0026thinsp;=\u0026thinsp;42). Based on COGs classification, many of those duplicated genes are part of the Replication, recombination and repair category. A manual inspection of the duplicated sequences in this category revealed 283 copies of a predicted gene, encoding a 716 to 790 amino acids protein with a mean identity of 99.66% (Additional file 5). Protein annotations from HMMER indicate that they all harbor a zinc-finger SWIM and a MULE related domain.\u003c/p\u003e \u003cp\u003eOnly a small proportion of duplicates (4.55%, n\u0026thinsp;=\u0026thinsp;81) are associated with cellular processes and signaling or metabolism. In total, 18.09% of the identified duplicates remain unclassified, having no match with a corresponding eggNOG description, GO term, KEGG pathway, Pfam domain or InterProScan ID.\u003c/p\u003e \u003cp\u003eAmong CEPs, the proportions of duplicate classes were different from other protein coding genes. A total of 12.43% (n\u0026thinsp;=\u0026thinsp;21) of CEPs showed duplication, with 52.38% (n\u0026thinsp;=\u0026thinsp;11) being classified as proximal and the remaining 47.62% (n\u0026thinsp;=\u0026thinsp;10) as dispersed. For proximal duplicates, there is a cluster on contig 11 of three CEPs (CEP11_63.94, CEP11_64.126, CEP11_65.8) with 100% sequence similarity. Still on the same contig, another CEP cluster showed four copies (CEP11_58.17, CEP11_33.146, CEP11_60.3, CEP11_63.133), of which two were proximal, and two dispersed. The predicted protein sequences of these four copies are 205 to 233 aa long and vary in similarity between 87.0% to 97.4%, possibly indicating a more ancient duplication. All these candidates are part of the RAPLH-like proteins, harbor the Y/F/WxC motif and are homologous to haustoria secreted candidate effector proteins of \u003cem\u003eG. cichoracearum\u003c/em\u003e isolates GcM1 and GcM3. Other proximal duplications appear on contig 31, with four CEPs of near equal length and sequence similarity ranging from 80.25\u0026ndash;95.12%.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eAssembly statistics of the \u003cem\u003eGolovinomyces ambrosiae\u003c/em\u003e genome\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAssembly metric\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eValue\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGenome size (bp)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e155\u0026nbsp;192 055\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of total fragments\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e87\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of contigs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e73\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of scaffolds\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAssembly N50 (bp)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3\u0026nbsp;548 631\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLongest fragment (bp)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12\u0026nbsp;043\u0026nbsp;756\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRepeat content (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e82.64\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGC content (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e41.57\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eProtein coding genes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6995\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCEPs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e169\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe assembled genome presented in this study is among the most complete PM genomes available and provides the first genomic insights into the cannabis PM pathogen. \u003cem\u003eGolovinomyces ambrosiae\u003c/em\u003e displays a very similar genome composition to other characterized powdery mildew species. Its size of 155.2 Mb is comparable to the 152.7 Mb of \u003cem\u003ePodosphaera xanthii\u003c/em\u003e YZU573 [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e56\u003c/span\u003e], 129.9 Mb of \u003cem\u003eG. magnicellulatus\u003c/em\u003e [57] and the predicted 173.8-221.8 Mb of \u003cem\u003eG. cichoracearum\u003c/em\u003e [51]. The number of genes in the assembly is also consistent with other published PM genomes, ranging mostly from 6046 to 9372 [58, 59, 60]. While it seems common for these genomes to be dominated by transposable elements, and specifically by LINEs and LTRs, \u003cem\u003eG. ambrosiae\u003c/em\u003e was found to possess the highest repetitive elements content among all published dicot PM genomes to date. With no evidence from the annotations of crucial genes required for repeat-induced point mutation (RIP), a mechanism widely employed among fungi for the disabling of TE [58, 61], their absence clearly supports the impressive invasion of repetitive regions in this cannabis pathogen.\u003c/p\u003e \u003cp\u003eThe abundance of TEs and the high number of genes implicated in their activity show remarkable genome plasticity in \u003cem\u003eG. ambrosiae\u003c/em\u003e. In general, TEs can have negative impacts on genetic activities due to their potential of replicating in functionally essential DNA regions; on the other hand, they can also be drivers of beneficial adaptations and novel genes [62, 63, 64]. Their expansion can be accentuated by a lack of sexual recombination, which is typical of powdery mildew fungi that rely on asexual reproduction, a dominant condition in the context of indoor cannabis production. Pathogenicity may also be affected by the activity of TEs. Indeed, effector coding genes present in repeat-rich regions are more susceptible to accumulating mutations, which often improve fitness of the pathogen [62].\u003c/p\u003e \u003cp\u003eThe detection of the high copy number of the MULE transposase and zinc SWIM finger domain-containing genes further demonstrates how dynamic the genome of this fungus may be. These characteristics are reminiscent of certain structural features related to the \u003cem\u003emudrA\u003c/em\u003e gene of maize, which encode the MURA protein, a putative transposase responsible of the transposition of the MULE-MuDR DNA transposons [65]. MULE-MuDRs are class II DNA transposons and have been shown in maize to be among the most mutagenic elements of all TEs [65]. Following a cut-and-paste transposition mechanism, it has been observed that they are not randomly inserted within the genome. In fact, when transposition occurs, MULE-MuDR have been found to have an insertion preference near the 5\u0026rsquo; ends of genes and with recombinationally active regions of the genome. Further analyses of those genes would be interesting to validate if they are truly active and indeed have a role in transposition. In consideration of these observations and the abundance of TEs composing the assembled genome, it is reasonable to suggest that there is likely substantial genetic diversity within the \u003cem\u003eG. ambrosiae\u003c/em\u003e species, potentially leading to isolates with differential virulence profiles.\u003c/p\u003e \u003cp\u003eThe genetic data regarding potential effector proteins provided by the assembly is also of great interest. Their abundance is in range with other identified CEPs of dicot PMs, particularly with the 159\u0026ndash;175 found in \u003cem\u003eG. cichoracearum\u003c/em\u003e [51]. Additionally, many candidate effectors of \u003cem\u003eG. ambrosiae\u003c/em\u003e harbor the Y/F/WxC motif, which is a characteristic shared among powdery mildew effectors [66]. This motif has also been found to be an important structural element for RALPH proteins [67]. For CEPs involved in pathogenicity, the focus is on proteins that are secreted by the haustoria during pathogen-host interactions. The presence of a signal peptide, the Y/F/WxC motif and protein homology with haustoria secreted candidate effectors are important features that facilitate the filtration of the most promising candidates based on this characteristic. Additionally, since all known PM \u003cem\u003eAvrs\u003c/em\u003e recognized by a corresponding host\u0026rsquo;s \u003cem\u003eR\u003c/em\u003e gene are linked to RALPH proteins [58], CEPs sharing sequence similarity with ribonucleases is another significant characteristic worthy of consideration. With identified sequences sharing either one or more of these key attributes, this study may provide a strong foundation for future identification of bona fide \u003cem\u003eAvr\u003c/em\u003e genes. Among those, the 14 CEPs sharing every of the \u003cem\u003eAvr\u003c/em\u003e-like features mentioned as depicted in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e appear to be the most promising. These candidates should be carefully studied in future transcriptomic analyses to first confirm their expression. Functional experiments should follow to confirm their role in pathogenicity. Of these CEPs, three were identified as proximal duplicates sharing perfect sequence identity. If these predicted candidates were found to be bona fide effectors implicated in virulence, this particular strain of \u003cem\u003eG. ambrosiae\u003c/em\u003e may have great fitness advantages by benefiting from an increased effector production [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e68\u003c/span\u003e]. In contrast, it could be a significant drawback if those sequences are related to avirulent phenotypes, requiring gain-of-virulence mutations in every copy to counteract resistance [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e69\u003c/span\u003e]. The other CEPs showing four non-identical copies are also interesting. If implicated in virulence, this could be a sign that these genes were subject to diversifying selection, potentially as a result of an arms race against the host [70].\u003c/p\u003e \u003cp\u003eDirect \u003cem\u003eAvr\u003c/em\u003e-R protein interactions have been validated in the \u003cem\u003eBlumeria graminis\u003c/em\u003e f. sp \u003cem\u003ehordei\u003c/em\u003e (\u003cem\u003eBgh\u003c/em\u003e) \u0026ndash; barley pathosystem [71, 72, 73]. In total, seven effector proteins of \u003cem\u003eBgh\u003c/em\u003e are recognized by their matching Mildew Locus a (MLA) gene, all part of the \u003cem\u003eMla\u003c/em\u003e locus, encoding multiple allelic NLRs. In all studies, the selection of top-ranking candidate effectors has been made through transcriptome-wide association studies (TWAS) and/or comparative genome analysis, with the goal of finding significant sequence polymorphisms matching the distinct phenotypes among numerous \u003cem\u003eBgh\u003c/em\u003e isolates. For all effectors, the avirulent and virulent genotypes were either differentiated by nonsynonymous SNPs, presence/absence of transcript, a transposon insertion in the gene sequence or splice site mutations leading to intron retention. The process of matching \u003cem\u003eAvr\u003c/em\u003e genes to their corresponding resistance protein in the host has also been performed in the wheat-powdery mildew \u003cem\u003eBlumeria graminis\u003c/em\u003e f. sp \u003cem\u003etritici\u003c/em\u003e pathosystem. In the host\u0026rsquo;s genome, the \u003cem\u003ePm3\u003c/em\u003e gene is the most diverse resistance gene studied, with 17 functional allelic variants identified to date. To those, three matching \u003cem\u003eAvr\u003c/em\u003e genes have been identified, and gain-of-virulence mutations have been attributed to the presence of a suppressor of avirulence gene (\u003cem\u003eSvr\u003c/em\u003e) and/or amino-acid polymorphisms [20, 74]. Other \u003cem\u003eAvrs\u003c/em\u003e associated with \u003cem\u003ePm1a\u003c/em\u003e, \u003cem\u003ePm2\u003c/em\u003e, \u003cem\u003ePm17\u003c/em\u003e, \u003cem\u003ePm8 and Pm60\u003c/em\u003e were also identified through QTL-mapping, genetic fine mapping, map-based cloning, genome-wide association studies, and avirulence depletion assay [69, 75, 76, 77, 78, 79]. In one of those studies, another gain-of-virulence mechanism such as complete deletion of the \u003cem\u003eAvr\u003c/em\u003e gene has been highlighted [76]. These types of studies and their findings are critical for the comprehension of plant-pathogen interactions regarding virulence.\u003c/p\u003e \u003cp\u003eGenetic characterisation of how resistance can hold or break down within certain pathogen populations is necessary for enlightened breeding strategies. In the context of cannabis powdery mildew, it is essential to extend the genomic profiling of multiple \u003cem\u003eG. ambrosiae\u003c/em\u003e strains, which will allow researchers to better understand the diversity of genetic determinants implicated in virulence. Future work on the identification and cloning of the \u003cem\u003eG. ambrosiae Avrs\u003c/em\u003e will allow functional validation against the currently identified \u003cem\u003eR\u003c/em\u003e genes in the \u003cem\u003eC. sativa\u003c/em\u003e germplasm. Subsequently, the unravelling of the distinct pathotypes will make it possible to confidently make use of the genetic resistance sources against powdery mildew. For wheat, it has been proposed to collectively create an interactive online \u003cem\u003eR\u003c/em\u003e-gene atlas, which would provide information on exploited \u003cem\u003eR\u003c/em\u003e gene and the corresponding \u003cem\u003eAvr\u003c/em\u003e genotypes against which they provide protection [80]. Additionally, population survey data would be included to track changes in the effectorome of pathogens from different locations, informing on potential resistance breakdown threats. Since research on the cannabis powdery mildew resistance is still in its infancy, it would be of great interest to generate such data for each new resistance source discovered before findings get scattered. At least, for each newly discovered \u003cem\u003eR\u003c/em\u003e gene in the host, equivalent efforts should be invested into deciphering the cognate \u003cem\u003eAvr\u003c/em\u003e gene of the pathogen. In this context, this study sets a strong foundation for such future characterisations of host-pathogen interactions, enabling the most efficient approach of counteracting the disease.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eaa: Amino acid\u003cbr\u003e\u0026nbsp;Avr: Avirulence gene\u003cbr\u003eBgh: \u003cem\u003eBlumeria graminis\u003c/em\u003e f. sp. \u003cem\u003ehordei\u003c/em\u003e\u003cbr\u003e\u0026nbsp;BIC: Bayesian Information Criterion\u003cbr\u003e\u0026nbsp;BUSCO: Benchmarking Universal Single-copy Orthologs\u003cbr\u003e\u0026nbsp;CEPs: Candidate effector proteins\u003cbr\u003e\u0026nbsp;Chr01: Chromosome 1\u003cbr\u003e\u0026nbsp;COGs: Clusters of Orthologous genes\u003cbr\u003e\u0026nbsp;HMW: High molecular weight\u003cbr\u003e\u0026nbsp;HR: Hypersensitive reaction\u003cbr\u003eITOL: Interactive Tree of Life\u0026nbsp;\u003cem\u003e\u003cbr\u003e\u0026nbsp;\u003c/em\u003eMLA: Mildew Locus a\u003cbr\u003e\u0026nbsp;MLO: Mildew resistance locus o\u003cbr\u003e\u0026nbsp;ONT: Oxford Nanopore Technologies\u003cbr\u003e\u0026nbsp;PM : Powdery mildew\u003cbr\u003e\u0026nbsp;R genes : Resistance genes\u003cbr\u003e\u0026nbsp;NLRs: Nucleotide-binding site and/or leucine-rich repeat domains\u003cbr\u003e\u0026nbsp;QTL: Quantitative trait locus\u003cbr\u003e\u0026nbsp;RALPH: RNase-Like Proteins associated with Haustoria\u003cbr\u003e\u0026nbsp;RIP: Repeat-induced point mutation\u003cbr\u003e\u0026nbsp;S: Susceptibility\u003cbr\u003e\u0026nbsp;Svr: Suppressor of avirulence\u003cbr\u003e\u0026nbsp;TWAS: Transcription-wide association studies\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics, approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003cstrong\u003e\u003cbr\u003e\u003c/strong\u003eThe genome assembly, Illumina paired-end reads, and Oxford Nanopore PromethION 2 reads have been deposited in the NCBI databases under BioProject accession number PRJNA1438116. The assembly is accessible at DDBJ/ENA/GenBank under the accession number JBWDLV000000000. The Illumina paired-end reads and Oxford Nanopore PromethION 2 have been deposited in NCBI Sequence Read Archive (SRA) under accession numbers SRR37663630 and SRR37669284. Data will be released publicly upon publication of the article.\u003c/p\u003e\n\u003cp\u003eReviewer link: https://dataview.ncbi.nlm.nih.gov/object/PRJNA1438116?reviewer=odqq7spltb4cjitk7vqmdq1dql\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was supported by the Greenhouse Research Chair in Plant Protection-MAPAQ-Premier Tech and the program NSERC-Alliance awarded to RRB. RCL is funded by Genome Canada, Genome Quebec, CIHR and NFRF. FAR was supported by an NSERC Scholarship.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eF.A.R. performed all bioinformatics workflows, analyses and wrote the initial draft and manuscript. Y.A. and B.C. provided guidance for the bioinformatics workflows and reviewed the manuscript. C.L. and F.A.R. performed the DNA extraction. R.R.B. wrote, reviewed and edited the manuscript. D.T. edited the manuscript. Funding acquisition and supervision was provided by R.R.B. and D.T. R.C.L. devised the Nanopore strategies in genome analysis and revised the manuscript. S.M., J.G. and G.Q.H.N. revised the manuscript, used the PromethION Solo 2 for WGS and bioinformatics analyses.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBelanger, R. R., Bushnell, W. R., Dik, A. J., \u0026amp; Carver, T. L. (2002). \u003cem\u003eThe Powdery Mildews: a Comprehensive Treatise\u003c/em\u003e. American Phytopathological Society Press.\u003c/li\u003e\n\u003cli\u003ePark, J. H., Choi, Y.-J., \u0026amp; Shin, H.-D. (2025). Revisiting Golovinomyces Species (Erysiphaceae) in Korea: Re-identification, New Records, and Description of Golovinomyces physalidis sp. nov. \u003cem\u003eMycobiology\u003c/em\u003e, \u003cem\u003e53\u003c/em\u003e(4), 450\u0026ndash;465. https://doi-org.acces.bibl.ulaval.ca/10.1080/12298093.2025.2517424\u003c/li\u003e\n\u003cli\u003eP\u0026eacute;pin, N., Punja, Z. K., \u0026amp; Joly, D. L. (2018). Occurrence of Powdery Mildew Caused by Golovinomyces cichoracearum sensu lato on Cannabis sativa in Canada. Plant Disease, 102(12), 2644. https://doi-org.acces.bibl.ulaval.ca/10.1094/PDIS-04-18-0586-PDN \u003c/li\u003e\n\u003cli\u003eScott, \u0026amp; Punja. (2021). Evaluation of disease management approaches for powdery mildew on Cannabis sativa L. (marijuana) plants. \u003cem\u003eCanadian Journal of Plant Pathology\u003c/em\u003e, \u003cem\u003e43\u003c/em\u003e(3), 394\u0026ndash;412. https://doi-org.acces.bibl.ulaval.ca/10.1080/07060661.2020.1836026 \u003c/li\u003e\n\u003cli\u003eHealth Canada (2025). \u003cem\u003eGood production practices guide for cannabis\u003c/em\u003e. Government of Canada. http://epe.lac-bac.gc.ca/100/201/301/weekly_acquisitions_list-ef/2019/19-30/publications.gc.ca/collections/collection_2019/sc-hc/H14-271-2018-fra.pdf \u003c/li\u003e\n\u003cli\u003eMaymon, M., Jerushalmi, S., \u0026amp; Freeman, S. (2020). First report of Golovinomyces cichoracearum sensu lato on Cannabis sativa in Israel. \u003cem\u003eNew Disease Reports\u003c/em\u003e, \u003cem\u003e42\u003c/em\u003e(1), 11. https://doi-org.acces.bibl.ulaval.ca/10.5197/j.2044-0588.2020.042.011 \u003c/li\u003e\n\u003cli\u003ePunja, Z. K., Collyer, D., Scott, C., Lung, S., Holmes, J., \u0026amp; Sutton, D. (2019). Pathogens and Molds Affecting Production and Quality of Cannabis sativa L. \u003cem\u003eFrontiers in Plant Science\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e, 1120. https://doi-org.acces.bibl.ulaval.ca/10.3389/fpls.2019.01120 \u003c/li\u003e\n\u003cli\u003eSzarka, D., Tymon, L., Amsden, B., Dixon, E., Judy, J., \u0026amp; Gauthier, N. (2019). First Report of Powdery Mildew Caused by Golovinomyces spadiceus on Industrial Hemp ( Cannabis sativa) in Kentucky. \u003cem\u003ePlant Disease\u003c/em\u003e, \u003cem\u003e103\u003c/em\u003e(7), 1773. https://doi-org.acces.bibl.ulaval.ca/10.1094/PDIS-01-19-0049-PDN \u003c/li\u003e\n\u003cli\u003eFarinas, C., \u0026amp; Hand, F. P. (2020). First Report of Golovinomyces spadiceus Causing Powdery Mildew on Industrial Hemp (Cannabis sativa) in Ohio. \u003cem\u003ePlant Disease\u003c/em\u003e, \u003cem\u003e104\u003c/em\u003e(10), 2727. \u003c/li\u003e\n\u003cli\u003eBrochu, A.-S., Labbe, C., B\u0026eacute;langer, R. R., \u0026amp; P\u0026eacute;rez-L\u0026oacute;pez, E. (2022). First Report of Powdery Mildew Caused by Golovinomyces ambrosiae on Cannabis sativa L. (marijuana) in Quebec, Canada. Plant Disease. https://doi-org.acces.bibl.ulaval.ca/10.1094/PDIS-02-22-0350-PDN \u003c/li\u003e\n\u003cli\u003eWiseman, M. S., Bates, T. A., Garfinkel, A. R., Ocamb, C. M., \u0026amp; Gent, D. H. (2021). First Report of Powdery Mildew Caused by Golovinomyces ambrosiae on Cannabis sativa in Oregon. \u003cem\u003ePlant Disease\u003c/em\u003e, \u003cem\u003e105\u003c/em\u003e(9), 2733. \u003c/li\u003e\n\u003cli\u003ePunja, Z. K. (2022). First report of the powdery mildew pathogen of hops, Podosphaeria macularis, naturally infecting cannabis (Cannabis sativa L., marijuana) plants under field conditions. \u003cem\u003eCanadian Journal of Plant Pathology\u003c/em\u003e, \u003cem\u003e44\u003c/em\u003e(2), 235\u0026ndash;249. https://doi-org.acces.bibl.ulaval.ca/10.1080/07060661.2021.1960424 \u003c/li\u003e\n\u003cli\u003eSalmon, E. S. (1900). A MONOGRAPH OF THE ERYSIPHACEAE. \u003cem\u003eMemoirs of the Torrey Botanical Club\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e, 1\u0026ndash;292. \u003c/li\u003e\n\u003cli\u003eQiu, P.-L., Liu, S.-Y., Bradshaw, M., Rooney-Latham, S., Takamatsu, S., Bulgakov, T. S., Tang, S.-R., Feng, J., Jin, D.-N., Aroge, T., Li, Y., Wang, L.-L., \u0026amp; Braun, U. (2020). Multi-locus phylogeny and taxonomy of an unresolved, heterogeneous species complex within the genus Golovinomyces (Ascomycota, Erysiphales), including G. ambrosiae, G. circumfusus and G. spadiceus. \u003cem\u003eBMC Microbiology\u003c/em\u003e, \u003cem\u003e20\u003c/em\u003e(1). https://doi-org.acces.bibl.ulaval.ca/10.1186/s12866-020-01731-9\u003c/li\u003e\n\u003cli\u003eBraun, U., \u0026amp; Cook, R. T. A. (2012). \u003cem\u003eTaxonomic manual of the Erysiphales (powdery mildews)\u003c/em\u003e. CBS-KNAW Fungal Biodiversity Centre.\u003c/li\u003e\n\u003cli\u003eMieslerov\u0026aacute;, B., Kitner, M., \u0026Scaron;enkyř\u0026iacute;k, J. B., Křiv\u0026aacute;nkov\u0026aacute;, T., Sedl\u0026aacute;řov\u0026aacute;, M., Braun, U., \u0026amp; Lebeda, A. (2024). The co-occurrence of two powdery mildew species on Cannabis sativa in the Czech Republic. Journal of Phytopathology, 172(3). https://doi-org.acces.bibl.ulaval.ca/10.1111/jph.13348 \u003c/li\u003e\n\u003cli\u003eBuirs, L., \u0026amp; Punja, Z. K. (2024). Integrated Management of Pathogens and Microbes in \u003cem\u003eCannabis sativa\u003c/em\u003e L. (Cannabis) under Greenhouse Conditions. \u003cem\u003ePlants\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(6), 786. https://doi.org/10.3390/plants13060786\u003c/li\u003e\n\u003cli\u003eRufus J Arkinrinlola, Hansen. (2023). Efficacy of Organic Fungicides Against Hemp Powdery Mildew Caused by Golovinomyces ambrosiae in a Greenhouse in Tennessee. \u003c/li\u003e\n\u003cli\u003eSirangelo, T. M. (2023). NLR- and mlo-Based Resistance Mechanisms against Powdery Mildew in Cannabis sativa. Plants (Basel, Switzerland), 13(1). https://doi-org.acces.bibl.ulaval.ca/10.3390/plants13010105 \u003c/li\u003e\n\u003cli\u003eBourras, S., Kunz, L., Xue, M., Praz, C. R., M\u0026uuml;ller, M. C., K\u0026auml;lin, C., Schl\u0026auml;fli, M., Ackermann, P., Fl\u0026uuml;ckiger, S., Parlange, F., Menardo, F., Schaefer, L. K., Ben-David, R., Roffler, S., Oberhaensli, S., Widrig, V., Lindner, S., Isaksson, J., Wicker, T., et al. (2019). The AvrPm3-Pm3 effector-NLR interactions control both race-specific resistance and host-specificity of cereal mildews on wheat. Nature Communications, 10(1), 2292. https://doi-org.acces.bibl.ulaval.ca/10.1038/s41467-019-10274-1 \u003c/li\u003e\n\u003cli\u003eSun, S., Fu, H., Wang, Z., Duan, C., Zong, X., Zhu, Z., \u0026amp; Singh, J. (2016). Discovery of a Novel er1 Allele Conferring Powdery Mildew Resistance in Chinese Pea (Pisum sativum L.) Landraces. \u003cem\u003ePlos One\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(1). https://doi-org.acces.bibl.ulaval.ca/10.1371/journal.pone.0147624 \u003c/li\u003e\n\u003cli\u003eMassonnet, M., Riaz, S., Pap, D., Figueroa-Balderas, R., Andrew Walker, M. \u0026amp; Cantu, D. (2022). The grape powdery mildew resistance loci Ren2, Ren3, Ren4D, Ren4U, Run1, Run1.2b, Run2.1, and Run2.2 activate different transcriptional responses to Erysiphe necator. Frontiers in Plant Science, 13. https://doi-org.acces.bibl.ulaval.ca/10.3389/fpls.2022.1096862 \u003c/li\u003e\n\u003cli\u003eJ\u0026oslash;rgensen, I. H. (1992). Discovery, characterization and exploitation of Mlo powdery mildew resistance in barley. Euphytica : International Journal of Plant Breeding, 63(1-2), 141\u0026ndash;152. https://doi-org.acces.bibl.ulaval.ca/10.1007/BF00023919 \u003c/li\u003e\n\u003cli\u003eMihalyov, P. D., \u0026amp; Garfinkel, A. R. (2021). Discovery and Genetic Mapping of PM1, a Powdery Mildew Resistance Gene in Cannabis sativa L. Frontiers in Agronomy, 3. https://doi-org.acces.bibl.ulaval.ca/10.3389/fagro.2021.720215 \u003c/li\u003e\n\u003cli\u003eP\u0026eacute;pin, N., Hebert, F. O., \u0026amp; Joly, D. L. (2021). Genome-Wide Characterization of the MLO Gene Family in Cannabis sativa Reveals Two Genes as Strong Candidates for Powdery Mildew Susceptibility. \u003cem\u003eFrontiers in Plant Science\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e, 729261. https://doi-org.acces.bibl.ulaval.ca/10.3389/fpls.2021.729261 \u003c/li\u003e\n\u003cli\u003eStack, G. M., Cala, A. R., Quade, M. A., Toth, J. A., Monserrate, L. A., Wilkerson, D. G., Carlson, C. H., Mamerto, A., Michael, T. P., Crawford, S., Smart, C. D., \u0026amp; Smart, L. B. (2024). Genetic Mapping, Identification, and Characterization of a Candidate Susceptibility Gene for Powdery Mildew in Cannabis sativa. \u003cem\u003eMolecular Plant-Microbe Interactions : MPMI\u003c/em\u003e, \u003cem\u003e37\u003c/em\u003e(1), 51\u0026ndash;61. https://doi.org/10.1094/MPMI-04-23-0043-R\u003c/li\u003e\n\u003cli\u003eSeifi, S., Leckie, K. M., Giles, I., O\u0026rsquo;Brien, T., MacKenzie, J. O., Todesco, M., Rieseberg, L. H., Baute, G. J., \u0026amp; Celedon, J. M. (2025). Mapping and characterization of a novel powdery mildew resistance locus (PM2) in Cannabis sativa L. Frontiers in Plant Science, 16, 1543229. https://doi-org.acces.bibl.ulaval.ca/10.3389/fpls.2025.1543229\u003c/li\u003e\n\u003cli\u003eKusch, S., Frantzeskakis, L., Lassen, B. D., K\u0026uuml;mmel, F., Pesch, L., Barsoum, M., Walden, K. D., \u0026amp; Panstruga, R. (2024). A fungal plant pathogen overcomes mlo-mediated broad-spectrum disease resistance by rapid gene loss. The New Phytologist, 244(3), 962\u0026ndash;979. https://doi-org.acces.bibl.ulaval.ca/10.1111/nph.20063\u003c/li\u003e\n\u003cli\u003eLo Presti, L., Lanver, D., Schweizer, G., Tanaka, S., Liang, L., Tollot, M., Zuccaro, A., Reissmann, S., \u0026amp; Kahmann, R. (2015). Fungal Effectors and Plant Susceptibility. Annual Review of Plant Biology, 66, 513\u0026ndash;545. https://doi-org.acces.bibl.ulaval.ca/10.1146/annurev-arplant-043014-114623\u003c/li\u003e\n\u003cli\u003eLinhan Li, Benjamin Collier, \u0026amp; Pietro Spanu. (2019). Isolation of Powdery Mildew Haustoria from Infected Barley. Bio-Protocol, 9(14). https://doi-org.acces.bibl.ulaval.ca/10.21769/BioProtoc.3299\u003c/li\u003e\n\u003cli\u003eMayjonade, B., Gouzy, J., Donnadieu, C., Pouilly, N., Marande, W., Callot, C., Langlade, N., \u0026amp; Mu\u0026ntilde;os, S. (2016). Extraction of high-molecular-weight genomic DNA for long-read sequencing of single molecules. BioTechniques, 61(4), 203\u0026ndash;205.\u003c/li\u003e\n\u003cli\u003eRusso, A., Mayjonade, B., Frei, D., Potente, G., Kellenberger, R. T., Frachon, L., Copetti, D., Studer, B., Frey, J. E., Grossniklaus, U., \u0026amp; Schl\u0026uuml;ter, P. M. (2022). Low-Input High-Molecular-Weight DNA Extraction for Long-Read Sequencing From Plants of Diverse Families. Frontiers in Plant Science, 13, 883897. https://doi-org.acces.bibl.ulaval.ca/10.3389/fpls.2022.883897\u003c/li\u003e\n\u003cli\u003eDe Coster, W., D\u0026apos;Hert, S., Schultz, D. T., Cruts, M., \u0026amp; Van Broeckhoven, C. (2018). NanoPack: Visualizing and processing long-read sequencing data. Bioinformatics, 34(15), 2666\u0026ndash;2669. https://doi.org/10.1093/bioinformatics/bty149 \u003c/li\u003e\n\u003cli\u003eLi, H. (2018). Minimap2: Pairwise alignment for nucleotide sequences. Bioinformatics, 34(18), 3094\u0026ndash;3100. https://doi.org/10.1093/bioinformatics/bty191 \u003c/li\u003e\n\u003cli\u003eKolmogorov, M., Yuan, J., Lin, Y., \u0026amp; Pevzner, P. A. (2019). Assembly of long, error-prone reads using repeat graphs. \u003cem\u003eNature Biotechnology, 37\u003c/em\u003e(5), 540\u0026ndash;546. \u003cstrong\u003ehttps://doi.org/10.1038/s41587-019-0072-8\u003c/strong\u003e \u003c/li\u003e\n\u003cli\u003eChen, S. (2023). Ultrafast one-pass FASTQ data preprocessing, quality control, and deduplication using fastp. \u003cem\u003eiMeta, 2\u003c/em\u003e(2), e107. \u003cstrong\u003ehttps://doi.org/10.1002/imt2.107\u003c/strong\u003e \u003c/li\u003e\n\u003cli\u003eWalker, B. J., Abeel, T., Shea, T., Priest, M., Abouelliel, A., Sakthikumar, S., Cuomo, C. A., Zeng, Q., Wortman, J., Young, S. K., \u0026amp; Earl, A. M. (2014). Pilon: An integrated tool for comprehensive microbial variant detection and genome assembly improvement. \u003cem\u003ePLoS ONE, 9\u003c/em\u003e(11), e112963. \u003cstrong\u003ehttps://doi.org/10.1371/journal.pone.0112963\u003c/strong\u003e \u003c/li\u003e\n\u003cli\u003eSim\u0026atilde;o, F. A., Waterhouse, R. M., Ioannidis, P., Kriventseva, E. V., \u0026amp; Zdobnov, E. M. (2015). BUSCO: Assessing genome assembly and annotation completeness with single-copy orthologs. \u003cem\u003eBioinformatics, 31\u003c/em\u003e(19), 3210\u0026ndash;3212. \u003cstrong\u003ehttps://doi.org/10.1093/bioinformatics/btv351\u003c/strong\u003e \u003c/li\u003e\n\u003cli\u003eBorowiec M. L. (2016). AMAS: a fast tool for alignment manipulation and computing of summary statistics. \u003cem\u003ePeerJ\u003c/em\u003e, \u003cem\u003e4\u003c/em\u003e, e1660. https://doi-org.acces.bibl.ulaval.ca/10.7717/peerj.1660\u003c/li\u003e\n\u003cli\u003eNguyen, L. T., Schmidt, H. A., von Haeseler, A., \u0026amp; Minh, B. Q. (2015). IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. \u003cem\u003eMolecular biology and evolution\u003c/em\u003e, \u003cem\u003e32\u003c/em\u003e(1), 268\u0026ndash;274. https://doi-org.acces.bibl.ulaval.ca/10.1093/molbev/msu300 \u003c/li\u003e\n\u003cli\u003eLetunic and Bork (2024) \u003cem\u003eNucleic Acids Res\u003c/em\u003e \u003cstrong\u003edoi: 10.1093/nar/gkae268\u003c/strong\u003e\u003c/li\u003e\n\u003cli\u003eFlynn, J. M., Hubley, R., Goubert, C., Rosen, J., Clark, A. G., Feschotte, C., \u0026amp; Smit, A. F. (2020). RepeatModeler2 for automated genomic discovery of transposable element families. Proceedings of the National Academy of Sciences of the United States of America, 117(17), 9451\u0026ndash;9457. https://doi.org/10.1073/pnas.1921046117\u003c/li\u003e\n\u003cli\u003eChen, N. (2004). Using RepeatMasker to identify repetitive elements in genomic sequences. \u003cem\u003eCurrent Protocols in Bioinformatics, 4\u003c/em\u003e, Unit 4.10. \u003c/li\u003e\n\u003cli\u003eHolt, C., \u0026amp; Yandell, M. (2011). MAKER2: an annotation pipeline and genome-database management tool for second-generation genome projects. \u003cem\u003eBMC bioinformatics\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e, 491. https://doi-org.acces.bibl.ulaval.ca/10.1186/1471-2105-12-491 \u003c/li\u003e\n\u003cli\u003eKeller, O., Kollmar, M., Stanke, M., \u0026amp; Waack, S. (2011). A novel hybrid gene prediction method employing protein multiple sequence alignments. \u003cem\u003eBioinformatics (Oxford, England)\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e(6), 757\u0026ndash;763. https://doi-org.acces.bibl.ulaval.ca/10.1093/bioinformatics/btr010 \u003c/li\u003e\n\u003cli\u003eAlmagro Armenteros, J.J., Tsirigos, K.D., S\u0026oslash;nderby, C.K. \u003cem\u003eet al.\u003c/em\u003e SignalP 5.0 improves signal peptide predictions using deep neural networks. \u003cem\u003eNat Biotechnol\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 420\u0026ndash;423 (2019). https://doi-org.acces.bibl.ulaval.ca/10.1038/s41587-019-0036-z\u003c/li\u003e\n\u003cli\u003eHallgren, J., Tsirigos, K. D., Pedersen, M. D., Almagro Armenteros, J. J., Marcatili, P., Nielsen, H., Krogh, A., \u0026amp; Winther, O. (2022). DeepTMHMM predicts alpha and beta transmembrane proteins using deep neural networks. \u003cem\u003ebioRxiv\u003c/em\u003e. \u003cstrong\u003ehttps://doi.org/10.1101/2022.04.08.487609\u003c/strong\u003e\u003c/li\u003e\n\u003cli\u003eSonah, H., Deshmukh, R. K., \u0026amp; B\u0026eacute;langer, R. R. (2016). Computational Prediction of Effector Proteins in Fungi: Opportunities and Challenges. \u003cem\u003eFrontiers in Plant Science\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e, 126. https://doi.org/10.3389/fpls.2016.00126\u003c/li\u003e\n\u003cli\u003eSperschneider J and Dodds PN (2021) EffectorP 3.0: prediction of apoplastic and cytoplasmic effectors in fungi and oomycetes. \u003cem\u003eMPMI\u003c/em\u003e. \u003c/li\u003e\n\u003cli\u003eSteinegger, M., \u0026amp; S\u0026ouml;ding, J. (2017). MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. \u003cem\u003eNature Biotechnology, 35\u003c/em\u003e(11), 1026\u0026ndash;1028. \u003c/li\u003e\n\u003cli\u003eWu, Y., Ma, X., Pan, Z., Kale, S. D., Song, Y., King, H., Zhang, Q., Presley, C., Deng, X., Wei, C. I., \u0026amp; Xiao, S. (2018). Comparative genome analyses reveal sequence features reflecting distinct modes of host-adaptation between dicot and monocot powdery mildew. \u003cem\u003eBMC genomics\u003c/em\u003e, \u003cem\u003e19\u003c/em\u003e(1), 705. https://doi-org.acces.bibl.ulaval.ca/10.1186/s12864-018-5069-z \u003c/li\u003e\n\u003cli\u003eCantalapiedra, C. P., Hern\u0026aacute;ndez-Plaza, A., Letunic, I., Bork, P., \u0026amp; Huerta-Cepas, J. (2021). eggNOG-mapper v2: Functional Annotation, Orthology Assignments, and Domain Prediction at the Metagenomic Scale. \u003cem\u003eMolecular biology and evolution\u003c/em\u003e, \u003cem\u003e38\u003c/em\u003e(12), 5825\u0026ndash;5829. https://doi-org.acces.bibl.ulaval.ca/10.1093/molbev/msab293 \u003c/li\u003e\n\u003cli\u003eJones P, Binns D, Chang H-Y, Fraser M, Li W, McAnulla C, McWilliam H, Maslen J, Mitchell A, Nuka G, et al. InterProScan 5: genome-scale protein function classification. Bioinformatics. 2014:30(9):1236\u0026ndash;1240. 10.1093/bioinformatics/btu03.\u003c/li\u003e\n\u003cli\u003eZimmermann, L., Stephens, A., Nam, S. Z., Rau, D., K\u0026uuml;bler, J., Lozajic, M., Gabler, F., S\u0026ouml;ding, J., Lupas, A. N., \u0026amp; Alva, V. (2018). A completely reimplemented MPI bioinformatics toolkit with a new HHpred server at its core. \u003cem\u003eJournal of Molecular Biology, 430\u003c/em\u003e(15), 2237\u0026ndash;2243. \u003cstrong\u003ehttps://doi.org/10.1016/j.jmb.2017.12.007\u003c/strong\u003e \u003c/li\u003e\n\u003cli\u003eWang, Y., Tang, H., DeBarry, J. D., Tan, X., Li, J., Wang, X., Lee, T., Jin, H., Marler, B., Guo, H., Kissinger, J. C., \u0026amp; Paterson, A. H. (2012). MCScanX: A toolkit for detection and evolutionary analysis of gene synteny and collinearity. \u003cem\u003eNucleic Acids Research, 40\u003c/em\u003e(7), e49. \u003cstrong\u003ehttps://doi.org/10.1093/nar/gkr1293\u003c/strong\u003e \u003c/li\u003e\n\u003cli\u003eWang, Z., Du, Y., Li, S., Xu, X., \u0026amp; Chen, X. (2023). A Complete Genome Sequence of \u003cem\u003ePodosphaera xanthii\u003c/em\u003e Isolate YZU573, the Causal Agent of Powdery Mildew Isolated from Cucumber in China. \u003cem\u003ePathogens (Basel, Switzerland)\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(4), 561. https://doi-org.acces.bibl.ulaval.ca/10.3390/pathogens12040561 \u003c/li\u003e\n\u003cli\u003eFarinas, C., Gluck-Thaler, E., Slot, J. C., \u0026amp; Peduto Hand, F. (2019). Whole-Genome Sequence of the Phlox Powdery Mildew Pathogen Golovinomyces magnicellulatus Strain FPH2017-1. \u003cem\u003eMicrobiology resource announcements\u003c/em\u003e, \u003cem\u003e8\u003c/em\u003e(36), e00852-19. https://doi-org.acces.bibl.ulaval.ca/10.1128/MRA.00852-19\u003c/li\u003e\n\u003cli\u003eKusch, S., Qian, J., Loos, A., K\u0026uuml;mmel, F., Spanu, P. D., \u0026amp; Panstruga, R. (2024). Long-term and rapid evolution in powdery mildew fungi. \u003cem\u003eMolecular ecology\u003c/em\u003e, \u003cem\u003e33\u003c/em\u003e(10), e16909. https://doi-org.acces.bibl.ulaval.ca/10.1111/mec.16909 \u003c/li\u003e\n\u003cli\u003eFrantzeskakis, L., N\u0026eacute;meth, M. Z., Barsoum, M., Kusch, S., Kiss, L., Takamatsu, S., \u0026amp; Panstruga, R. (2019). The \u003cem\u003eParauncinula polyspora\u003c/em\u003e Draft Genome Provides Insights into Patterns of Gene Erosion and Genome Expansion in Powdery Mildew Fungi. \u003cem\u003emBio\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(5), e01692-19. https://doi-org.acces.bibl.ulaval.ca/10.1128/mBio.01692-19 \u003c/li\u003e\n\u003cli\u003eGa\u0026ntilde;\u0026aacute;n, L., White, R. A., 3rd, Friesen, M. L., Peever, T. L., \u0026amp; Amiri, A. (2020). A Genome Resource for the Apple Powdery Mildew Pathogen \u003cem\u003ePodosphaera leucotricha\u003c/em\u003e. \u003cem\u003ePhytopathology\u003c/em\u003e, \u003cem\u003e110\u003c/em\u003e(11), 1756\u0026ndash;1758. https://doi-org.acces.bibl.ulaval.ca/10.1094/PHYTO-05-20-0158-A \u003c/li\u003e\n\u003cli\u003eSpanu, P. D. et al. (2010). Genome expansion and gene loss in powdery mildew fungi reveal tradeoffs in extreme parasitism. \u003cem\u003eScience (New York, N.Y.)\u003c/em\u003e, \u003cem\u003e330\u003c/em\u003e(6010), 1543\u0026ndash;1546. https://doi-org.acces.bibl.ulaval.ca/10.1126/science.1194573 \u003c/li\u003e\n\u003cli\u003eM\u0026ouml;ller, M., \u0026amp; Stukenbrock, E. H. (2017). Evolution and genome architecture in fungal plant pathogens. \u003cem\u003eNature reviews. Microbiology\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(12), 756\u0026ndash;771. https://doi-org.acces.bibl.ulaval.ca/10.1038/nrmicro.2017.76 \u003c/li\u003e\n\u003cli\u003eNottensteiner, M., Zechmann, B., McCollum, C., \u0026amp; H\u0026uuml;ckelhoven, R. (2018). A barley powdery mildew fungus non-autonomous retrotransposon encodes a peptide that supports penetration success on barley. \u003cem\u003eJournal of experimental botany\u003c/em\u003e, \u003cem\u003e69\u003c/em\u003e(15), 3745\u0026ndash;3758. https://doi-org.acces.bibl.ulaval.ca/10.1093/jxb/ery174 \u003c/li\u003e\n\u003cli\u003eSabelleck, B., \u0026amp; Panstruga, R. (2018). Novel jack-in-the-box effector of the barley powdery mildew pathogen?. \u003cem\u003eJournal of experimental botany\u003c/em\u003e, \u003cem\u003e69\u003c/em\u003e(15), 3511\u0026ndash;3514. https://doi-org.acces.bibl.ulaval.ca/10.1093/jxb/ery192 \u003c/li\u003e\n\u003cli\u003eLisch, D. (2014). Mutator and MULE transposons. Microbiology Spectrum, 3(2), Article MDNA3-0032-2014. https://doi.org/10.1128/microbiolspec.MDNA3-0032-2014\u003c/li\u003e\n\u003cli\u003eGodfrey, D., B\u0026ouml;hlenius, H., Pedersen, C., Zhang, Z., Emmersen, J., \u0026amp; Thordal-Christensen, H. (2010). Powdery mildew fungal effector candidates share N-terminal Y/F/WxC-motif. BMC genomics, 11, 317. https://doi-org.acces.bibl.ulaval.ca/10.1186/1471-2164-11-317 \u003c/li\u003e\n\u003cli\u003eCao, Y., K\u0026uuml;mmel, F., Logemann, E., Gebauer, J. M., Lawson, A. W., Yu, D., Uthoff, M., Keller, B., Jirschitzka, J., Baumann, U., Tsuda, K., Chai, J., \u0026amp; Schulze-Lefert, P. (2023). Structural polymorphisms within a common powdery mildew effector scaffold as a driver of coevolution with cereal immune receptors. Proceedings of the National Academy of Sciences of the United States of America, 120(32), e2307604120. https://doi-org.acces.bibl.ulaval.ca/10.1073/pnas.2307604120\u003c/li\u003e\n\u003cli\u003ePedersen, C., Ver Loren van Themaat, E., McGuffin, L. J., Abbott, J. C., Burgis, T. A., Barton, G., Bindschedler, L. V., Lu, X., Maekawa, T., Wessling, R., Cramer, R., Thordal-Christensen, H., Panstruga, R., \u0026amp; Spanu, P. D. (2012). Structure and evolution of barley powdery mildew effector candidates. \u003cem\u003eBMC genomics\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e, 694. https://doi-org.acces.bibl.ulaval.ca/10.1186/1471-2164-13-694\u003c/li\u003e\n\u003cli\u003eM\u0026uuml;ller, M. C., Kunz, L., Schudel, S., Lawson, A. W., Kammerecker, S., Isaksson, J., Wyler, M., Graf, J., Sotiropoulos, A. G., Praz, C. R., Manser, B., Wicker, T., Bourras, S., \u0026amp; Keller, B. (2022). Ancient variation of the \u003cem\u003eAvrPm17\u003c/em\u003e gene in powdery mildew limits the effectiveness of the introgressed rye \u003cem\u003ePm17\u003c/em\u003e resistance gene in wheat. \u003cem\u003eProceedings of the National Academy of Sciences of the United States of America, 119\u003c/em\u003e(30), Article e2108808119. \u003cstrong\u003ehttps://doi.org/10.1073/pnas.2108808119\u003c/strong\u003e\u003c/li\u003e\n\u003cli\u003eLiu, L., Xu, L., Jia, Q., Pan, R., Oelm\u0026uuml;ller, R., Zhang, W., \u0026amp; Wu, C. (2019). Arms race: diverse effector proteins with conserved motifs. \u003cem\u003ePlant signaling \u0026amp; behavior\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(2), 1557008. https://doi-org.acces.bibl.ulaval.ca/10.1080/15592324.2018.1557008 \u003c/li\u003e\n\u003cli\u003eLu, X., Kracher, B., Saur, I. M. L., Bauer, S., Ellwood, S. R., Wise, R., Yaeno, T., Maekawa, T., \u0026amp; Schulze-Lefert, P. (2016). Allelic barley MLA immune receptors recognize sequence-unrelated avirulence effectors of the powdery mildew pathogen. \u003cem\u003eProceedings of the National Academy of Sciences of the United States of America, 113\u003c/em\u003e(36), E6486\u0026ndash;E6495. \u003cstrong\u003ehttps://doi.org/10.1073/pnas.1612947113\u003c/strong\u003e \u003c/li\u003e\n\u003cli\u003eSaur, I. M. L., Bauer, S., Kracher, B., Lu, X., Franzeskakis, L., M\u0026uuml;ller, M. C., Sabelleck, B., K\u0026uuml;mmel, F., \u0026amp; Panstruga, R. (2019). Multiple pairs of allelic MLA immune receptor-powdery mildew AVRA effectors argue for a direct recognition mechanism. \u003cem\u003eeLife, 8\u003c/em\u003e, Article e44471. \u003cstrong\u003ehttps://doi.org/10.7554/eLife.44471\u003c/strong\u003e\u003c/li\u003e\n\u003cli\u003eBauer, S., Yu, D., Lawson, A. W., Saur, I. M. L., Frantzeskakis, L., Kracher, B., Logemann, E., Chai, J., Maekawa, T., \u0026amp; Schulze-Lefert, P. (2021). The leucine-rich repeats in allelic barley MLA immune receptors define specificity towards sequence-unrelated powdery mildew avirulence effectors with a predicted common RNase-like fold. \u003cem\u003ePLoS Pathogens, 17\u003c/em\u003e(1), Article e1009223. \u003cstrong\u003ehttps://doi.org/10.1371/journal.ppat.1009223\u003c/strong\u003e \u003c/li\u003e\n\u003cli\u003eBourras, S., McNally, K. E., Ben-David, R., Parlange, F., Roffler, S., Praz, C. R., Oberhaensli, S., Menardo, F., Stirnweis, D., Frenkel, Z., Schaefer, L. K., Fl\u0026uuml;ckiger, S., Treier, G., Herren, G., Korol, A. B., Wicker, T., \u0026amp; Keller, B. (2015). Multiple avirulence loci and allele-specific effector recognition control the \u003cem\u003ePm3\u003c/em\u003e race-specific resistance of wheat to powdery mildew. \u003cem\u003eThe Plant Cell, 27\u003c/em\u003e(10), 2991\u0026ndash;3012. \u003cstrong\u003ehttps://doi.org/10.1105/tpc.15.00171\u003c/strong\u003e \u003c/li\u003e\n\u003cli\u003eHewitt, T., M\u0026uuml;ller, M.C., Moln\u0026aacute;r, I., Mascher, M., Holu\u0026scaron;ov\u0026aacute;, K., \u0026Scaron;imkov\u0026aacute;, H., Kunz, L., Zhang, J., Li, J., Bhatt, D., Sharma, R., Schudel, S., Yu, G., Steuernagel, B., Periyannan, S., Wulff, B., Ayliffe, M., McIntosh, R., Keller, B., Lagudah, E. and Zhang, P. (2021), A highly differentiated region of wheat chromosome 7AL encodes a \u003cem\u003ePm1a\u003c/em\u003e immune receptor that recognizes its corresponding \u003cem\u003eAvrPm1a\u003c/em\u003e effector from \u003cem\u003eBlumeria graminis\u003c/em\u003e. New Phytol, 229: 2812-2826. https://doi-org.acces.bibl.ulaval.ca/10.1111/nph.17075\u003c/li\u003e\n\u003cli\u003ePraz, C. R., Bourras, S., Zeng, F., S\u0026aacute;nchez-Mart\u0026iacute;n, J., Menardo, F., Xue, M., Yang, L., Roffler, S., B\u0026ouml;ni, R., Herren, G., McNally, K. E., Ben-David, R., Parlange, F., Oberhaensli, S., Fl\u0026uuml;ckiger, S., Sch\u0026auml;fer, L. K., Wicker, T., Yu, D., \u0026amp; Keller, B. (2017). AvrPm2 encodes an RNase-like avirulence effector which is conserved in the two different specialized forms of wheat and rye powdery mildew fungus. \u003cem\u003eNew Phytologist, 213\u003c/em\u003e, 1301\u0026ndash;1314. \u003cstrong\u003ehttps://doi.org/10.1111/nph.14372\u003c/strong\u003e \u003c/li\u003e\n\u003cli\u003eManser, B., Koller, T., Praz, C. R., Roulin, A. C., Zbinden, H., Arora, S., Steuernagel, B., Wulff, B. B. H., Keller, B., \u0026amp; S\u0026aacute;nchez-Mart\u0026iacute;n, J. (2021). Identification of specificity-defining amino acids of the wheat immune receptor Pm2 and powdery mildew effector AvrPm2. \u003cem\u003eThe Plant Journal, 106\u003c/em\u003e(4), 993\u0026ndash;1007. \u003cstrong\u003ehttps://doi.org/10.1111/tpj.15214\u003c/strong\u003e \u003c/li\u003e\n\u003cli\u003eKunz, L., Sotiropoulos, A. G., Graf, J., Razavi, M., Keller, B., \u0026amp; M\u0026uuml;ller, M. C. (2023). The broad use of the Pm8 resistance gene in wheat resulted in hypermutation of the AvrPm8 gene in the powdery mildew pathogen. \u003cem\u003eBMC biology\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e(1), 29. https://doi-org.acces.bibl.ulaval.ca/10.1186/s12915-023-01513-5 \u003c/li\u003e\n\u003cli\u003eKunz, L., Jigisha, J., Menardo, F., Sotiropoulos, A. G., Zbinden, H., Zou, S., Tang, D., H\u0026uuml;ckelhoven, R., Keller, B., \u0026amp; M\u0026uuml;ller, M. C. (2025). Avirulence depletion assay: Combining R gene-mediated selection with bulk sequencing for rapid avirulence gene identification in wheat powdery mildew. \u003cem\u003ePLoS pathogens\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e(1), e1012799. https://doi-org.acces.bibl.ulaval.ca/10.1371/journal.ppat.1012799 \u003c/li\u003e\n\u003cli\u003eHafeez, A. N., Arora, S., Ghosh, S., Gilbert, D., Bowden, R. L., \u0026amp; Wulff, B. B. H. (2021). Creation and judicious application of a wheat resistance gene atlas. \u003cem\u003eMolecular plant\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(7), 1053\u0026ndash;1070. https://doi-org.acces.bibl.ulaval.ca/10.1016/j.molp.2021.05.014\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-genomics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"gics","sideBox":"Learn more about [BMC Genomics](http://bmcgenomics.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/gics","title":"BMC Genomics","twitterHandle":"#BMCGenomics","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Powdery mildew, Genome, Cannabis sativa, Genetic resistance, Effector, Avirulence","lastPublishedDoi":"10.21203/rs.3.rs-9162455/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9162455/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eCannabis powdery mildew, caused by the fungal pathogen \u003cem\u003eGolovinomyces ambrosiae\u003c/em\u003e, poses a significant threat to licensed producers as its presence on marketable products can compromise product innocuity. While there is a focus in research for resistance genes within the \u003cem\u003eCannabis sativa\u003c/em\u003e germplasm, there is a lack of genetic information regarding the infecting agent, preventing validation of their effectiveness against different populations as the plant-pathogen interaction most likely follows a gene-for-gene relationship. In this paper, we assembled the first \u003cem\u003eG. ambrosiae\u003c/em\u003e genome, providing insights into its genomic content and potential virulence determinants. The assembly was made using a hybrid approach, combining Oxford Nanopore Technologies long reads and Illumina short reads.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThe resulting 155.2 Mb genome is composed of 73 contigs, 13 scaffolds and has a completeness score of 97.5%. Subsequent analysis highlighted the substantial transposable elements content of the pathogen, occupying 82.64% of its genomic composition. Prediction of protein-coding genes revealed 6995 highly confident gene models, including 169 candidate effector proteins. Among the latter, we highlighted 14 candidates sharing key characteristics of confirmed effectors in other powdery mildew pathosystems such as the presence of signal peptides, RALPH-like domains, and Y/F/WxC motifs.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eThe result of this study provides valuable resources for future identification of avirulence genes within the \u003cem\u003eG. ambrosiae\u003c/em\u003e species, responsible of conferring resistance when encoded effectors are recognized by a cognate host\u0026rsquo;s resistance genes. Those findings will lead to guided breeding strategies and, ultimately, the selection of cultivars adapted to the pathogen\u0026rsquo;s virulence profile.\u003c/p\u003e","manuscriptTitle":"Hybrid genome assembly of the cannabis powdery mildew agentGolovinomyces ambrosiae uncovers important resources for deciphering virulence factors","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-16 17:13:51","doi":"10.21203/rs.3.rs-9162455/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-22T08:05:28+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-19T18:38:58+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-17T12:35:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"124067635127659701903146431619959813761","date":"2026-04-11T11:23:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"188475353049841951163421759496589223066","date":"2026-04-10T09:32:25+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-08T18:37:01+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-02T20:29:48+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-31T13:29:09+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Genomics","date":"2026-03-31T13:22:37+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-genomics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"gics","sideBox":"Learn more about [BMC Genomics](http://bmcgenomics.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/gics","title":"BMC Genomics","twitterHandle":"#BMCGenomics","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"eedf4168-17ae-4392-a27c-c3192beddc39","owner":[],"postedDate":"April 16th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-05-14T10:11:12+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-16 17:13:51","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9162455","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9162455","identity":"rs-9162455","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}