IMA Genome-F 4: Draft genome sequences of Chrysoporthe austroafricana, Diplodia scrobiculata, Fusarium nygamai, Leptographium lundbergii, Limonomyces culmigenus, Stagonosporopsis tanaceti, and Thielaviopsis punctulata.

The genomes of Chrysoporthe austroafricana, Diplodia scrobiculata, Fusarium nygami, Leptographium lundbergii, Limonomyces culmigenus, Stagonosporopsis tanaceti, and Thielaviopsis punctulata are presented in this genome announcement. These seven genomes are from endophytes, plant pathogens and economically important fungal species. The genome sizes range from 26.6 Mb in the case of Leptographium lundbergii to 44 Mb for Chrysoporthe austroafricana. The availability of these genome data will provide opportunities to resolve longstanding questions regarding the taxonomy of species in these genera, and may contribute to our understanding of the lifestyles through comparative studies with closely related organisms.

Diplodia scrobiculata was originally considered to be a morphotype (B) of D. sapinea (syn. Diplodia pinea) until marker technology made it possible to recognise it as a discrete species (de Wet ). The fungus is an endophyte and latent pathogen of Pinus species known especially from the USA and some countries of western Europe (Palmer 1987, Blodgett & Stanosz 1997, Stanosz , Burgess ). It was recently reported from the Southern Hemisphere for the first time when it was found in South Africa (Bihon ).

Diplodia scrobiculata is a weak pathogen that has been isolated from various Pinus species and some other conifers where it often co-exists with D. sapinea (de Wet ). A population genetics study on this fungus has shown that isolates are genetically diverse where it is native, and that there is strong geographic isolation between populations from different regions of North America (Burgess ). With the exception of a few studies on the epidemiology and population structure, very little is known of the biology of D. scrobiculata.

Recently, the genome sequence of D. sapinea became available for public use (van der Nest ). Together with this and other sequences for species of Botryosphaeriaceae (Phillips ) the availability of a genome sequence for D. scrobiculata would promote detailed biological studies and provide the basis for interesting genome comparisons including for example those considering pathogenicity factors.

SEQUENCED STRAIN

South Africa: Mpumanlanga: Sappi, isol. ex Pinus patula, Oct. 2008, W. Bihon (CMW 30223 = CBS139796; PREM 61246 – dried culture).

NUCLEOTIDE SEQUENCE ACCESSION NUMBER

The Whole Genome Shotgun projects have been deposited at DBJ/EMBL/GenBank under the accessions LAEG00000000. The version described in this paper is version LAEG01000000.

METHODS

DNA from a single spore culture of Diplodia scrobiculata (CMW30223 = CBS139796) was extracted and sequenced using the Illumina: HiSeq genome analyser platform at Inqaba Biotechnology Industries, South Africa. Reads were subjected to the necessary sequence quality analysis and those of less than 30 bases and poor quality regions of reads were discarded. Remaining reads were assembled into a draft genome using CLC Genomic de novo assembler 7.0, with the minimum contig size set to 500 bases (CLCbio, Aarhus, Denmark). Gene predictions were done using AUGUSTUS (Stanke ) based on the gene models for Aspergillus oryzae (http://bioinf.uni-greifswald.de/webaugustus/prediction). Completeness of the genome was evaluated using the Core Eukaryotic Genes Mapping Approach (CEGMA) analysis (Parra ). Contigs of ≥ 500 bases were submitted to the genome database of NCBI.

RESULTS AND DISCUSSION

The draft nuclear sequence of D. scrobiculata isolate CMW30223 = CBS139796 resulted in an estimated genome size of 35.85 megabases (Mb). De novo assembly using CLC Genomics produced a total of 3789 contigs with N50 of 16 kilobases (Kb) and maximum contig size of 119715 bases. Mapping to the more conserved set of 248 Core Eukaryotic Genes (CEGs) using CEGMA pipeline (Parra ), more than 95.2 % of the genome sequence was complete. Putative gene prediction using AUGUSTUS (Stanke ) predicted 13624 genes.

Diplodia scrobiculata has a genome size similar to that of D. sapinea (36.9Mb) and Alternaria alternata (33 Mb) (van der Nest , http://genome.jgi.doe.gov/). It has a similar number of genes to A. alternata (13469) but more than the most closely D. sapinea (13020). Although, D. scrobiculata is commonly found in association with D. sapinea, it is not clear why it is mainly confined to the Northern Hemisphere and that it has not moved globally as its sibling species has. It is also not known why it is considerably less pathogenic than D. sapinea (de Wet ). Public access to the genome sequence of D. scrobiculata together with other related species will facilitate answering these questions as well as others, such as those relating to the association of endophytic latent pathogens and their tree hosts.

Authors: W. Bihon, M.J. Wingfield, B. Slippers, and B.D. Wingfield

Species of Chrysoporthe (family Cryphonectriaceae) are important tree pathogens (Wingfield 2003), occasionally causing serious canker disease on Myrtales (Gryzenhout et al. 2004). Chrysoporthe austroafricana is mainly found in southern African countries, where it affects susceptible plantation-grown Eucalyptus species as well as native Syzygium species (Wingfield , Myburg , Roux , Heath , Nakabonge , Vermeulen ). Symptoms on Eucalyptus often include cankers at the base and root collars of trees, causing wilt and subsequent death of young trees. Older trees show swollen basal cankers which girdle the stem and cause breakage during high winds.

The genus Chrysoporthe also includes C. cubensis (South and Central America) and C. deuterocubensis (south-east Asia) (Hodges , 1979, Myburg , Rodas , Gryzenhout , van der Merwe ). Five other species of Chrysoporthe are currently recognized: C. doradensis (Ecuador) (Gryzenhout ), C. syzygiicola and C. zambiensis (Zambia) (Chungu ), and C. inopina and C. hodgesiana (Columbia) (Gryzenhout , 2006).

Previous studies on Chrysoporthe species have focused on their taxonomy (Gryzenhout ), population genetics (Nakabonge , van der Merwe , Vermeulen ), and geographical distribution (Myburg , Rodas , Heath , Nakabonge , Vermeulen ). However, in order to further explore the genetics and evolution of these species in relation to their mating systems, population biology, phylogenetics and potential biotechnological applications, we have determined the whole genome sequence of C. austroafricana isolate CMW2113. This genome resource will not only broaden the scope of research on the eight Chrysoporthe species, but also other fungi in the Cryphonectriaceae.

South Africa: KwaMbonambi, 1989, M.J. Wingfield (culture CMW2113 = CBS 112916; PREM 58023 – dried culture).

The Chrysoporthe austroafricana isolate CMW2113 Whole Genome Shotgun project was deposited in GenBank with accession number JYIP00000000. The version described here is JYIP01000000.

Genomic DNA was extracted from C. austroafricana isolate CMW2113. Genome sequencing was done at the Agricultural Research Council (ARC, South Africa) using the Illumina MiSeq paired-end protocol (Illumina, CA). Additionally, three independent genomic DNA samples were sequenced using the single-read IonTorrent protocol (Life Technologies Carlsbad, CA) at the University of Pretoria. These three batches consisted of two batches sequenced using a 100-base sequencing kit, and one batch using a 300-base kit. CLC Genomics Workbench v. 7.5.1 (CLCbio, Aarhus, Denmark) was used to assemble the paired-end MiSeq sequences after quality assessment. Additional scaffolding was performed with the unused MiSeq and IonTorrent data using SSPACE 2.0 (Boetzer et al. 2011). Protein coding gene models were annotated de novo using AUGUSTUS v. 3.0.3 with Fusarium graminearum as the model organism (Stanke & Morgenstern 2005). Lastly, the completeness of the genome was assessed using the occurrence of core eukaryotic orthologous genes, as implemented in the CEGMA v. 2.4 software (Parra ).

The assembled and scaffolded C. austroafricana genome was 44 669 169 bases in size, including gaps. The coverage calculated from 6 416 contigs was 40X, with an N50 value of 61 627 bases. Further, the GC content was estimated at 53.9 %. Analysis of the genome completeness using CEGMA showed that 94.35 % of 445 core orthologous genes were present. The AUGUSTUS gene prediction pipeline predicted a total of 13 484 protein coding gene models.

The total C. austroafricana genome size, predicted at 44.6 Mb is slightly larger than that of Cryphonectria parasitica (43.9Mb; 11 184 genes) (http://genome.jgi.doe.gov/Crypa2 /Crypa2.info.html), which is the only other member of Cryphonectriaceae for which genome information is available. The evolutionary significance of these differences is yet to be determined, including whether these variations played a role in the diversification of the species. The C. austroafricana genome size and number of predicted protein coding genes were also compared to model ascomycetes such as Neurospora crassa (39.9 Mb, 10 082 genes; Galagan ), Podospora anserina (36Mb, 10 545 genes) (EspagneĪ ) and Magnaporthe grisea, (40.3Mb, 11 109 genes; Dean ), all of which were smaller with fewer predicted protein coding genes.

The continued availability of fungal genomes provides an opportunity to broaden the scope of research in fungal biology, taxonomy, and evolution. This is especially true for species of Chrysoporthe, since the lack of genomic resources has limited the scope of previous studies. The C. austroafricana WGS is an invaluable resource that will certainly advance our knowledge of the biology of Chrysoporthe species.

Authors: A.M. Kanzi, B.D. Wingfield, M.J. Wingfield, E.T. Steenkamp, and N.A. van der Merwe

The Fusarium fujikuroi species complex (FFSC, previously referred to as the Gibberella fujikuroi species complex) includes a range of important pathogens as well as species of human and animal health concern because of their ability to produce highly toxic compounds (Leslie & Summerell 2006, Kvas ). To allow genomic comparisons of these important fungi, the genomes of a growing number of FFSC species are available, e.g. F. verticillioides, F. circinatum, F. fujikuroi, and F. mangiferae (Fusarium Comparative Sequencing Project, Wingfield et al. 2012, Wiemann ). These fungi are, respectively, pathogens of maize, pine, rice, and mango and are by no means representative of the FFSC diversity known to be associated with cultivated plants (Leslie & Summerell 2006, Kvas ). This is true, not only in terms of plant host, but also other biological properties (e.g. morphology, mycotoxicology) typically linked with this complex. For example, none of the four species for which whole genome sequences are capable of forming chlamydospores (Kvas ). In fact, the ability to produce these thick-walled vegetative structures is not a common FFSC feature and only a few species have been shown to have this capability (Leslie & Summerell 2006, Kvas ).

The chlamydospore formers in the FFSC include F. dlamini, F. napiforme, F. nygami, F. acutatum, F. pseudoanthophilum, F. udum, and F. xylarioides (O’Donnell , 2000).

Among these, the important wilt pathogens, F. udum, F. xylarioides (Booth 1971), and F. nagamai have probably been most extensively studied. The aim of this study was to sequence the whole genome of F. nagamai. The species represents one of the so-called mating populations or biological species of the complex (Klaasen & Nelson 1996) and is often associated with sorghum and other field crops. The species may be dominant in native grassland soils, especially in hot and arid locations (Sangalang ). Fusarium nygami has also been demonstrated to be the causal agent of basal stalk and root rot of grain sorghum (Trimboli & Burgess 1983), while certain isolates have been reported to cause disease on broad beans, asparagus, cotton, maize, millet, and rice (Leslie & Summerell 2006). Overall, the genome sequence of this species would allow for more extensive genomic comparisons across the FFSC and facilitate in depth explorations of the genetic basis for the diverse biological characters found in this complex.

USA: Pennsylvania: laboratory strain, progeny of a sexual cross between two F. nygamai strains (M-1564 x M-2370) isolated from soil in South Africa and soil in Australia. MAT-1 mating type tester strain for mating population G (Klaasen & Nelson 1996), June 1996, J.A. Klaasen & P.E. Nelson (MRC8546, CMW12847, KSU G-05111, CMWF1209, M-7492, K-148, PREM – 61277 dried culture).

The Fusarium nygamai genome sequence data has been deposited at DDBJ/EMBL/GenBank under the accession number LBNR 0000 0000. The first version is described in this paper.

Genomic DNA was isolated from 1-wk-old single-spore cultures incubated on half strength potato dextrose agar (PDA) (Biolab Diagnostics, Wadeville, South Africa) at 25 °C as described previously (Groenewald ). One mate-pair and two pair-end libraries with insert sizes of 3000 bp, 250–350bp and 560–650 bp, respectively, were prepared and subjected to Illumina HiSeq 2000 sequencing at Fasteris (Genèva, Switzerland). After exclusion of poor quality reads using CLC Genomics Workbench v. 8.0 (CLCbio, Aarhus, Denmark), sequences were assembled into scaffolds using ABySS v. 1.5.2 (Simpson ) and gapped regions within scaffolds were closed using GapFiller v1.11 (Boetzer & Pirovano 2012). The completeness of the genome assembly was evaluated using CEGMA v. 2.5 (Parra ) and BUSCO using conserved orthologous gene sets specific to fungi (Simão ). Putative open reading frames (ORFs) were predicted using WebAUGUSTUS (Hoff & Stanke 2013) together with the gene models for F. graminearum (Fusarium Comparative Sequencing Project http//www.broad.mit.edu) and F. circinatum (Wingfield )

Assembly of the draft genome of Fusarium nygamai yielded a genome size of 51 615 029 bp with an average coverage of 333x. The genome was assembled into 409 scaffolds ranging in size from 200 bp to 5 319 023 bp with an N50 scaffold length of 2 761 656 bp and an average GC content of 47.46 %. Of the 409 contigs, 15 scaffolds were larger than 1 000 000 bases in size and represented approximately 78 % of the complete genome sequence. Based on the occurrence of a set of genes shared by all eukaryotes (Parra ) the F. nygamai genome is 96.37 % complete and 99 % complete based on the conserved fungal gene sets. The genome of F. nygamai is predicted to contain 15 780 putative ORFs.

The genome of F. nygamai is larger than those of F. verticillioides (41.8 Mb), F. circinatum (44.3 Mb), F. fujikuroi (43.9 Mb), and F. mangiferae (45.6 Mb) (Wiemann et al. 2013). The genome size, however, is similar to that of F. solani (51.3 Mb) (Wiemann et al. 2013) with the predicted number of ORFs in agreement to not only F. solani but also the other smaller genomes of species in the FFSC (Wiemann et al. 2013). Preliminary BLAST analysis also suggests that the F. nygamai isolates sequenced in this study contains twelve chromosomes. The twelfth chromosome appears to be dispensable in some members of the genus Fusarium and was found to be absent from a laboratory strain of F. circinatum (van der Nest ). The availability of this genome, should therefore not only assist in future phylogenomic studies of the FFSC species which would allow for a better understanding of the overall evolution of this complex, but could also assist in a better understanding of the role and importance of dispensable chromosomes in the FFSC.

Authors:

Leptographium lundbergii, described in 1927 as a cause of blue stain in spruce in Sweden (Lagerberg ), is the type species of Leptographium s. lat. This is an ascomycete genus residing in Ophiostomatales that currently accommodates more than 100 species (de Beer & Wingfield 2013). Most species of Leptographium s. lat. are associates of bark and root-infesting beetles found on conifers and are best known for their ability to cause sap stain in recently killed trees (Jacobs & Wingfield 2001). Species of Leptographium s. lat. are mostly considered to be saprobes, but a small number are important tree pathogens (Harrington 1993, Jacobs & Wingfield 2001). Phylogenetically, the pathogens are closely related to those species that have a saprobic lifestyle (Linnakoski , de Beer & Wingfield 2013).

The genomes of some Leptographium s.lat. associated with tree disease have been sequenced and characterized. These include those of Grosmannia clavigera and Leptographium procerum (DiGuistini , van der Nest et al. 2014). The availability of a genome sequence of a saprobic species such as L. lundbergii would make possible opportunities for genome-level comparisons between pathogenic and non-pathogenic species, and possibly provide insights into how pathogenic species cause disease. The aim of this study was thus to generate genome data of L. lundbergii, a saprobe and importantly the type species of Leptographium s. lat.

Norway: Pinus sylvestris, 1989, H. Roll-Hansen (culture CMW 2190 = CBS 138716; PREM 61278 – dried culture)

The genomic sequence of Leptographium lundbergii (CMW 2190 = CBS 138716) has been deposited at DDBJ/EMBL/GenBank under the accession LDEF00000000. The version described in this paper is version LDEF01000000.

Genomic DNA was extracted from a single conidium culture of L. lundbergii CMW 2190 (CBS 138716) following the method described by Duong . Genomic DNA was submitted for Illumina Hiseq sequencing at the Genome Centre, University of California at Davis (CA). Two insert libraries (350 bp and 530 bp average insert size) were prepared and 100 bp pair-end sequencing was carried out on the Illumina HiSeq 2000 platform. Reads with low quality were filtered out and 15 bases of sequence were trimmed from 5’ ends of all reads. De-novo assembly was carried out with reads after filtering and trimming using the CLC genomic work bench v. 8.0.1 (CLCBio, Aarhus, Denmark). The quality and completeness of the assembled genome was estimated by calculating CEGMA score (Parra ). The protein coding genes were predicted using the MAKER genome annotation and curation pipeline (Cantarel ). Firstly, RepeatScout (Price ) was used to search for repetitive regions of the genome. These regions were then masked from the genome using RepeatMasker (Smit et al. 2013–15). For protein coding gene prediction, SNAP (Korf 2004), Augustus (Stanke & Waack 2003) and GeneMark (Lukashin & Borodovsky 1998) gene prediction programmes were used. Two rounds of self-training were conducted for SNAP using proteins from Grossmania clavigera (DiGuistini ), Ophiostoma ulmi (Khoshraftar ), O. novo-ulmi (Comeau ), O. piceae (Haridas ), Magnaporthe grisea (Dean ), and Neurospora crassa (Galagan ) as protein homology evidence. The species gene model of M. grisea was used for Augustus prediction and the HMM file generated with a separated run was used for GeneMark. InterProScan (Zdobnov & Apweiler 2001) was used to screen the predicted proteins for the presence of PfamA domains to use as additional support for resulting gene models. The completeness of the genome was estimated using CEGMA (Parra , 2009) and BUSCO (Simão ). The fungal mating type locus of the sequenced isolate was identified and characterized by using Blast against known mating type genes from closely related species (Duong , Tsui ).

About 22 million reads with an average read length of 84.5 bp were obtained after quality filtering and trimming. De-novo assembly using the CLC Genomics Workbench resulted in 412 scaffolds with over 500 bp in size. The N50 value was 212 Kb and the longest scaffold was just over 1 Mb. The coverage across the whole assembly was around 70 times. The estimated genome size of L. lundbergii is 26.6 Mb, smaller than the closely related species, L. procerum (28.6 Mb) and G. clavigera (29.8 Mb) (DiGuistini , van der Nest ). The mean GC content of the L. lundbergii genome was 56.8 %, which is a slightly higher than that of G. clavigera (53.4 %), and L. procerum (54.77 %). The assembled L. lundbergii genome had a CEGMA completeness score of 95.97 % when calculated from the complete gene set and 97.18 % when calculated from both partial and complete gene sets. Assessment of the completeness of genome using 1438 BUSCO groups for fungi resulted in BUSCO values of C:97%[D:5.5%],F:2.0%,M:0.4%,n:1438 (C:complete [D:duplicated], F:fragmented, M:missed, n:genes). These results from CEGMA and BUSCO indicated that the assembled genome covered most of the organism’s gene space. MAKER predicted a total of 9547 protein coding genes (MAKER’s max build), 8035 of which were supported by protein and/or PfamA domain evidence (MAKER’s standard build) (Campbell ). Leptographium lundbergii had a gene density of 1/2.78 kb and the mean GC of the coding regions was 54 % (minimum of 27.5 % and maximum of 70.2 %). Of all suggested gene models from MAKER, 2070 were single exon (21.7 %) and 7477 gene models were multi-exon (78.3 %). The median exon length was 544 bp and the median intron length was 126 bp.

The sequenced L. lundbergii isolate harboured the MAT1-2 idiomorph with a typical MAT1-2-1 gene and a truncated version of MAT1-1-1 gene. This is similar to the MAT1-2 idiomorphs characterized in a number of closely related species of Leptographium s. lat. (Duong , Tsui ). The draft genome of L. lundbergii represents a useful resource that will enable future comparative genomic studies to be conducted between saprophytic and pathogenic species of Leptographium and related fungi. As the type species of the genus, the genome sequence of L. lundbergii will also be valuable for future systematic studies on the relationships between members of this genus and other species in the Ophiostomatales.

Authors: T.A. Duong, M.J. Wingfield, Z.W. de Beer, and B.D. Wingfield

The genus Stagonosporopsis from Didymellaceae (de Gruyter ) includes several economically important plant pathogens such as S. cucurbitacearum (syn. Didymella bryoniae), the cause of gummy blight of cucurbits, and S. andigena and S. crystalliniformis; destructive quarantine pathogens of potato and tomato, respectively (Aveskamp , de Gruyter ). Stagonosporopsis tanaceti (formerly known as Phoma ligulicola var. inoxydabilis) is the cause of ray blight, the most damaging disease of pyrethrum (Tanacetum cinerariifolium) in Australia (Pethybridge & Wilson 1998, Vaghefi ). The fungus was first detected in the Australian pyrethrum fields in 1995, and soon became a major threat, causing devastating epidemics in 1999, which led to losses of to 100 % (Pethybridge ). The origin and global distribution of S. tanaceti is unknown but it is believed to have close evolutionary affinity to S. chrysanthemi and S. inoxydabilis, which infect various Asteraceae in Europe and the USA (Stevens 1907, van der Aa ).

The homothallic S. chrysanthemi and S. inoxydabilis have well-described sexual morphs in Didymella (de Gruyter ). However, no sexual morph has been detected in S. tanaceti, and it is believed to reproduce mainly asexually in pyrethrum fields based on the detection of only one mating-type gene in the population as well as significant linkage disequilibrium of neutral genetic markers (Pethybridge , Vaghefi , b). It causes deformation and dieback of the leaves, stem lesions, and flower blight, which can result in flower death and complete yield loss if not controlled with multiple fungicide applications (Pethybridge ).

Stagonosporopsis tanaceti is also known to live as an endophyte on symptomless plant tissue, becoming pathogenic when environmental conditions are conducive to disease development (unpublished data). However, the infection process and pathogenicity mechanisms of S. tanaceti are not yet fully understood. Many plant pathogenic Dothideomycetes are known to produce toxins that contribute to host-specific or non-specific pathogenicity (Daub & Ehrenshaft 2000, Friesen ). Indeed, S. chrysanthemi produces phytotoxic compounds upon infection of chrysanthemum (Schadler & Bateman 1974, 1975) but these compounds have not been described in the S. tanaceti-pyrethrum pathosystem. Availability of genomic data will provide a powerful tool for understanding the mechanisms of pathogenicity and infection process of S. tanaceti through studying the genes underlying such mechanisms.

Moreover, genome sequence data provides a valuable resource for detection of the genes responsible for sensitivity to fungicides. Disease management strategies for ray blight in Australian pyrethrum fields are heavily reliant upon fungicides (Pethybridge , 2008). However, the use of fungicides with a high risk of resistance development (e.g. strobilurins and succinate dehydrogenase inhibitors; FRAC 2013) is a major concern to the durability of disease management. Identifying genomic regions associated with fungicide sensitivity will enable rapid screening of S. tanaceti populations for characterizing temporal changes in resistant allele frequencies in response to specific selection pressures. A de novo genome assembly of S. tanaceti is presented and made publically available to facilitate understanding the genetic mechanisms underlying the biology and evolution of this important plant pathogen.

SEQUENCED STRAINS

Australia: Tasmania: northern Tasmania, Scottsdale, from Tanacetum cinerariifolium (pyrethrum), 2004, S.J. Pethybridge (CBS H-20947 – holotype; CBS 131484 – culture ex-holotype); from T. cinerariifolium, 2004, S.J. Pethybridge (CBS 131485); from T. cinerariifolium, 2004, S.J. Pethybridge (TAS 55503); from T. cinerariifolium, 1998, S.J. Pethybridge (DAR 70020).

This Whole Genome Shotgun project has been deposited at DDBJ/EMBL/GenBank under the accession JUDZ00000000. The version described in this paper is version JUDZ01000000.

The genome of Stagonosporopsis tanaceti (a pooled DNA sample of four isolates) was sequenced through the Illumina platform (100 bp paired-end reads) at the Savannah River Ecology Laboratory (SREL, University of Georgia, GA). This yielded 29 395 716 paired reads, totalling 5.8 Gb. Trimmomatic v.0.22 (http://www.usadellab.org/cms/?page=trimmomatic) was used for quality trimming with a 4-base long sliding window, a minimum quality value of 30, and minimum length of 30 bases. A total of 21 553 834 paired reads (73.32 %) and 5 519 748 single reads (18.78 %) were retained for assembly after quality trimming (7.9 % discarded). Partial genome assembly was conducted using SOAPdenovo v. 1.05 (Li ) with the kmer length parameter tested between 21 and 51 in 2 bases increments. The optimal kmer resulting in the maximal N50 length occurred at k = 35. The completeness of the genome was assessed using the Core Eukaryotic Genes Mapping Approach (CEGMA v. 1.0) (Parra ). Gene prediction was conducted in Maker v. 2.31.7 (Cantarel ); a genome annotation pipeline that combine ab initio gene prediction and evidence-based annotations, suitable for novel genomes with no training data available. Only contigs with a length of ≥ 500 bp were used for annotation. A preliminary annotation utilized ab initio gene prediction programs SNAP (Korf 2004) and Augustus v. 2.5.5 (Stanke ) as well as evidence-based gene prediction using the ESTs and protein sequences of Didymella exigua genome (http://genome.jgi-psf.org/Didex1/Didex1.home.html). The resulting annotation was used to produce a hidden-markov-model (HMM) profile for S. tanaceti, which was further refined with a second stage of SNAP training. The refined HMM file was used for the final annotation (Cantarel ).

The draft genome had a total assembly size of 40.8 Mbp with a genome coverage of at least 100×, N50 value of 98 157 bp and maximum contig size of 112 607 bp. All contigs with a length of ≥ 200 bp (9 834 contigs) were submitted to the genome database of NCBI. Mapping to the conserved set of 248 Core Eukaryotic Genes (CEGs) using CEGMA (Parra ) estimated the genome to be at least 94 % complete, as the analysis revealed hits for 234 CEGs (94.35 %) with complete match and 244 with partial match (98.39 %). Ab initio gene prediction using trained SNAP identified 17 219 Open Reading Frames (ORFs).

The estimated genome size of S. tanaceti (40.8 Mb) is similar to the genome size of multiple Dothideomycetes, including Pyrenophora teres (41.95 Mb), Zymoseptoria tritici (39.7 Mb), and Stagonospora nodorum (37.2 Mb) (Hane , Ellwood , Goodwin et al. 2011). This was much larger than the genome of Didymella exigua, which has a predicted size of 34.39 Mb (http://genome.jgi-psf.org/Didex1/Didex1.i nfo.html).

The draft genome of S. tanaceti, has been used for the detection of the mating-type locus (Vaghefi ) and development of a microsatellite library (Vaghefi ). To the best of our knowledge, this is the first reported draft genome of the genus Stagonosporopsis and provides a useful genomic resource for further genome-enabled studies into plant-microbe interactions.

Authors: N. Vaghefi, P.K. Ades, S.J. Pethybridge, and P.W.J. Taylor

Basidiomycete fungi in the order Corticiales are an ecologically diverse group of organisms that derive nutrients through pathogenic associations with plants, lichenicolous symbioses, or saprophytically from dead organic matter (Lawrey et al. 2008). Two plant-associated genera in the Corticiales, Limonomyces and Laetisaria, are particularly well known for their vivid red and pink coloration (Stalpers & Loerakker 1982). Morphology is the primary method used to discriminate between members of Limonomyces and Laetisaria (Stalpers & Loerakker 1982). However, based on phylogenetic analyses of nuclear and mitochondrial ribosomal DNA, both Laetisaria and Limonomyces appear to be polyphyletic, and are inadequately described from morphological data alone (Diederich ).

Members of the genus Limonomyces are associated with grasses in several genera, including Agrostis, Cynodon, Festuca, Lolium, and Poa (Smiley et al. 2005). The genus includes two species: L. roseipellis and L. culmigenus (Stalpers & Loerakker 1982). Limonomyces roseipellis and L. culmigenus closely resemble one another, except L. culmigenus develops larger basidiospores and basidia, which have two sterigmata (Stalpers & Loerakker 1982). In natural environments, L. roseipellis is often found in association with Laetisaria fuciformis (Smiley et al. 2005). Both L. roseipellis and Laetisaria fuciformis are weakly pathogenic to grasses under conducive environmental conditions, causing diseases known as pink patch and red thread, respectively (Smiley et al. 2005). In managed turf grasses, red thread and pink patch diseases are unsightly, therefore turf grass breeders actively work to incorporate resistance to these pathogens in their cultivars (Bonos ).

Unlike L. roseipellis and Laetisaria fuciformis, the lifestyle of Limonomyces culmigenus is poorly known, with just a few living samples of the fungus collected from grasses in England and Canada during the past five decades. The fungus has been described as a parasite of Dactylis glomerata, primarily growing within the host tissue and producing narrow pink sporophores (Stalpers & Loerakker 1982). References to L. culmigenus as a causal agent of turfgrass pink patch are sometimes found, but no credible evidence demonstrating pathogenicity or disease is available (Farr & Rossman 2015). To enable research on this enigmatic fungus, we generated a draft genome assembly from L. culmigenus CBS 661.85. This draft assembly provides the first genomic resource for a member of the family Corticiaceae.

Canada: British Columbia: isolated from dead grass culm, 1985, J.R. Bandoni (CBS 661.85; BPI 893193 – dried culture).

NUCLEOTIDE ACCESSION NUMBERS

The Whole Genome Shotgun project of the CBS 661.85 genome has been deposited in NCBI GenBank under the accession LCTX00000000, with version LCTX00000000 described in this paper. Nucleotide sequences for homeodomain (HD1, HD2), pheromone receptor (STE3), and mitochondrial intermediate peptidase precursor (MIP) genes are deposited in NCBI GenBank under accession numbers KR779915−KR779922.

Genomic DNA was extracted and purified as previously described (Malapi-Wight ) from a natural isolate of Limonomyces culmigenus, CBS 661.85. An Illumina TruSeq PCR-Free DNA sequencing library (Illumina, San Diego, CA) was constructed and quantified using the Qubit 2.0 fluorometer (Life Technologies, Grand Island, NY) and LabChipXT DNA 750 (Caliper Life Sciences, Hopkinton, MA), then sequenced on an Illumina MiSeq instrument using a 600-cycle sequencing cartridge (Illumina, Inc). Reads were processed and assembled using the CLC Genomics Workbench v. 7.6.1 software (CLC Bio, Boston, MA). After adapter removal, reads were trimmed to remove low quality sequences (Phred quality score <0.05) and reads shorter than 15 or longer than 600 bases. A de novo genome sequence was assembled based on de Bruijn graphs using a kmer=61, bubble size of 50, with contigs >500 bp retained. Assembly summary statistics were calculated using CLC Genomics Workbench, PRINSEQ v. 0.20.4 (Schmieder and Edwards 2011) and QUAST β (Gurevich ). The number of core eukaryotic genes present in the assembly was assessed using CEGMA (Parra ) through the iPLANT Discovery Environment interface (https://de.iplantcollaborative.org/de/) and gene predictions were obtained using AUGUSTUS v. 3.0.2 with Coprinopsis cinerea gene models (Stanke ). Identification of protein ortholog families was performed using OrthoMCL (Li ). Read depth of coverage and variant detection relative to the genome assembly were generated in CLC Genomics Workbench, with variant calls restricted to regions with a minimum of 10x read coverage and >90% probability, modeled as a diploid. Scaffolds were analyzed for large-scale similarities with other scaffolds in the assembly by performing a reciprocal BLASTn analysis of scaffolds >1000 bp, then determining the average depth of reads across: (a) 100 pairs of scaffolds where a matching scaffold was identified from the assembly; and (b) 100 individual scaffolds without a matching scaffold in the assembly. Putative secreted proteins lacking GPI-anchor sites and transmembrane domains were identified using SECRETOOL (Cortazar ). Putative secondary metabolite clusters were identified using antiSMASH (Blin ), and CAZymes were predicted using dbCAN (Yin ) from the assemblies of L. culmigenus and three additional Agaricomycotina species downloaded from the Joint Genome Institute Mycocosm Genome Portal (http://genome.jgi.doe.gov/programs/fungi/index.jsf) for comparative purposes. The mitochondrial genome was identified by performing a tBLASTn search of the L. culmigenus CBS 661.85 assembly with the complete mitochondrial genome of Rhizoctonia solani AG8 (Scaffold 77; 140 kb; Hane ). Genes were identified from the mitochondrial genome using MITOS (Bernt ) under genetic code four. Transfer RNA was identified from the CBS 661.85 mitochondrial genome using tRNAscan-SE v. 1.21 (Lowe & Eddy 1997), with the source designated as mitochondrial and the genetic code as yeast mitochondrial (Schattner ). The mitochondrial assembly was compared with R. solani Rhs1AP using the promer, showcoords, mapview, and mummerplot functions in MUMmer v. 3.23 (Kurtz ). All BLAST searches in this work were performed with an E-value threshold of 1.0E-5.

The nuclear genome assembly of Limonomyces culmigenus CBS 661.85 is 36.1 Mb (30 843 264 paired-end reads; N50=9 365, L50=706). The average read depth of coverage is 211-fold, with a GC content of 57.8 %. The assembly is organized into 8 889 scaffolds of 500 bp or greater in length, with the largest scaffold measuring 250 757 bp (av.= 4 065 bp). The assembly size of CBS 661.85 is consistent with the genomes sequences of other published Agaricomycetes, such as Schizophyllum commune (38.5 Mb; Ohm ), C. cinerea (37 Mb; Stajich ) and Phanerochaete chrysosporium (35.1 Mb; Martinez ). The nuclear assembly is approximately 91.5 % complete, based on the presence of a portion of the ultra-conserved eukaryotic gene set (Parra ). The high depth of read coverage and relatively fragmented assembly suggest that CBS 661.85 has a repetitive genome.

Although a comprehensive analysis of the genome characteristics of L. culmigenus CBS 661.85 is beyond the scope of the present report, the predicted dikaryotic nature of this fungus indicates that the 36.1 Mb assembly is likely an overestimate of the haploid genome size. Three features of the CBS 661.85 assembly are consistent with a dikaryotic organism, as proposed for L. culmigenus based on morphological observations (Stalpers & Loerakker 1982). Variant detection assessment identified 935 heterozygous variants (92.6 % single nucleotide polymorphisms; SNPs) across 674 scaffolds (7.6 % of the total scaffolds), suggesting a high level of similarity between the two nuclei in these regions. All but eight of the identified variants are present at a 1:1 ratio relative to the assembly, consistent with the presence of two nuclei. We also observed an almost two-fold difference in the average depth of read coverage between scaffolds identified as unique in the assembly (average read coverage 173.4x), versus the read coverage calculated from scaffolds where a matching scaffold is present in the assembly (average read coverage 83.4x). Duplicated copies of the genes associated with the mating type locus also support the presence of two unique nuclei in the assembly. The A mating-type locus is represented by two scaffolds containing pairs of divergently transcribed genes containing homeodomain transcription factors (HD1 and HD2) alongside the mitochondrial intermediate peptidase precursor (MIP) gene on scaffolds 1965 and 2914. Pairwise nucleotide identities for the HD genes are 83.5 % and 67.6 % (HD1 and HD2, respectively), and the two MIP genes share 95.8 % identity. Two highly divergent copies of the STE3 pheromone receptor from the B mating type locus are present in the genome (scaffolds 5234 and joined_contig_68), with only 38.0% shared identity at the nucleotide level. The A and B mating type loci are located on different scaffolds, but due to the relatively small size of these scaffolds, we could not determine if the separation is due to fragmentation of the assembly, or if the loci occupy different locations in the genome. Therefore, it is still unknown whether L. culmigenus CBS 661.85 employs a bipolar or tetrapolar mating strategy.

A total of 16 394 putative coding regions were identified from the L. culmigenus CBS 661.85 genome assembly, with an average gene length of 1139 bp and a mean gene density of 454 genes/Mb. Proteins in the predicted proteome ranges from 16−6 262 amino acids in length (av.= 379). Comparison with available basidiomycete fungal genomes shows a general agreement between the predicted gene numbers of L. culmigenus and other Agaricomycetes; i.e. Laccaria bicolor, Coprinopsis cinerea and Schizophyllun commune with 20 614, 13 342, and 13 210 predicted coding regions respectively (Stajich ). Ortholog families were identified for 75.6 % of the L. culmigenus proteome.

Putative effector genes that function in the host infection process are normally identified from genome datasets as proteins that contain secretion signals, are small and cysteine rich, and are lineage specific. There are 1027 predicted secreted proteins in the L. culmigenus CBS 661.85 genome assembly, making up 8 % of the predicted proteome. Predicted amino acid sequence lengths within the secretome range from 36−1833 aa (av. = 300.5 aa). Within the secretome, 601 proteins are classified as small (SSPs; ≤300 aa), and from among these, 125 are cysteine-rich (relative cysteine content of ≥ 5 %; SSCPs). Several of the multi-copy SSCPs are positioned adjacent to one another, and may have arisen due to tandem duplication events. Fifty-two of the SSCPs are predicted as members of 23 known ortholog families, and include enzymes that would facilitate breakdown of host tissue, such as cellulases, peptide hydrolases, exo-beta-1,3-glucanases, and aspartic and serine endopeptidases. The predicted effecterome consists of the remaining 87 SSCPs that do not have similarity to any known ortholog family and produce no BLASTp hits in NCBI GenBank. Twenty-nine of the predicted effectorome proteins are present as single copies in the genome assembly, and 58 of these proteins are members of multi-copy groups of 2–9 proteins.

As with the other three Agaricomycotina species surveyed, the plant pathogen Rhizoctonia solani, the saprophytic jelly fungus Tremella mesenterica, and the saprophytic fungus Coprinopsis cinerea, the L. culmigenus CBS 661.85 genome assembly contains a minimal cohort of gene clusters involved in the biosynthesis of secondary metabolites, including one non-ribosomal peptide synthase, five polyketide synthase clusters and eight terpene clusters (Table 1). There are 809 CAZyme domains in the genome of CBS 661.85. Comparison of the CAZyme profiles between L. culmigenus and three other Agaricomycotina species shows that L. culmigenus possesses a cohort of CAZymes that is most similar to the plant pathogenic R. solani. In particular, both L. culmigenus and R. solani have a greatly expanded cohort of glycoside hydrolases (GH), relative to the two saprophytic species (1.5x–1.6x; Table 1). The increase in GH motifs is due to an overall increase of all GH families, and not attributable to expansion of any one family. As expected for a fungus that utilizes grass as a nutritional substrate, there are only nine pectate lyase (PL) motifs present in the L. culmigenus genome. This reduction in PL motifs is in stark contrast with the genome of R. solani AG-1 1B, where 92 PL motifs were predicted. To our knowledge, the abundance of PL motifs in R. solani AG-1 1B is the largest repertoire reported for any fungus to date, and may play a role in R. solani’s wide host range and adaptability.

Table 1.

Summary of the draft genome sequence of Limonomyces culmigenus (CBS 661.85) and published genome sequences for three basidiomycete fungi (Stajich , Floudas , Wibberg ).

Species	Limonomyces culmigenus CBS 661.85	Rhizoctonia solani AG-1 1B	Tremella mesenterica Fries v1.0	Coprinopsis cinerea 7#130
Summary Data
Total sequence length (Mb)	36.1	47.7	28.6	37.5
Number of scaffolds	8889	1600	45	13
GC (%)	57.8	48.1	45.7	51.7
Predicted gene models	16394	15157	8313	13393
Predicted secreted proteins	1027	876	213	776
Predicted CAZymes
Total CAZymes	809	944	282	661
Auxiliary activity	86	132	95	105
Carbohydrate binding	125	130	12	122
Carbohydrate esterase	128	141	9	115
Glycoside hydrolase	355	348	78	212
Glycosyltransferase	106	101	84	89
Pectate lyase	9	92	4	18
Predicted Secondary Metabolite Clusters
Total SM Clusters	23	13	4	14
Type I polyketide synthetases (PKSs)	3	0	0	1
Type III PKSs	2	0	0	0
Nonribosomal peptide synthetases	1	2	0	1
Terpene clusters	8	8	3	9
Hglks	0	0	0	0

The mitochondrial genome of L. culmigenus CBS 661.85 was identified from scaffold 72, which measures 67.4 kb and has a G+C content of 27 % (average read coverage 3 505x). Based on tRNA and gene content, this scaffold likely represents the complete mitochondrial genome. Twenty-seven tRNAs were identified from the mitochondrial scaffold, coding for all 20 amino acids. Transfer RNAs for five amino acids were duplicated (Arg, Phe, Ser=2 tRNAs; Leu, Met=3 tRNAs). The standard set of 14 conserved mitochondrial protein coding genes (Bullerwell & Lang 2005) was identified, sometimes in multiple copies, including all three ATP synthase subunits (ATP6, ATP8, ATP9), apocytochrome b (COB), all three cytochrome oxidase subunits (COX1, COX2, COX3), and all seven NADH dehydrogenase subunits (NAD1, NAD2, NAD3, NAD4, NAD4L, NAD5, NAD6). As reported from R. solani Rhs1AP, most mitochondrial genes and tRNAs were found on the same strand (Losada ); with the exception of NAD1 and six of the tRNAs. Overall, 53 proteins were predicted, including the 14 conserved proteins, two predicted homing endonucleases with GIY_YIG domains, seven predicted homing endonucleases with LAGLIDADG domains, two rRNAs and 13 hypothetical proteins. Core mitochondrial proteins from L. culmigenus CBS 661.85 shared 79–100 % amino acid identity to their homologues in R. solani Rhs1AP (Losada ). However, gene order comparisons showed almost no synteny between the mitochondria of these two species (data not shown). In general, the mitochondrial genome of L. culmigenus CBS 661.85 is average relative to the sequenced mitochondria observed from the other 32 basidiomycete fungi reported on NCBI GenBank (http://www.ncbi.nlm.nih.gov/genomes/ORGANELLES/organelles.html, accessed 05/05/2015), with respect to size (av. = 72.2 kb), G+C content (av. = 28.3), tRNA content (av. = 24.5 tRNAs) and gene content (av. = 32.6 genes).

The draft genome sequence of L. culmigenus CBS 661.85 is the first Corticiaceae genomic resource publically available. This dataset will provide opportunities to resolve longstanding questions regarding taxonomy for organisms in this group, and may contribute to our understanding of the lifestyle of L. culmigenus of other Corticiaceae through comparative studies with closely related organisms.

Authors: M. Malapi-Wight, L.A. Beirn, D. Veltri, and J.A. Crouch

Date palm (Phoenix dactylifers) is an important subsistence crop in arid regions due to its ability to grow under harsh environmental conditions such as high temperature and drought. Nevertheless, the ideal conditions for its growth and production are also favourable to fungal diseases, such as black scorch disease. Thielaviopsis punctulata (previously known as Ceratocystis radicola) is the causative agent of various plant-associated diseases, including black scorch (Suleman , Al-Naemi ), rhizosis (Linde & Smit 1999), basal trunk root rot (Kile , Mohammadi 1992, Suleman ), and sudden death disease in both the USA and South Africa (Bliss 1941, Linde & Smit 1999). Additionally, it infects other plant hosts causing leaf and root rot in pineapple (Kile 1993) and fruit rot in lemon (Mirzaee & Mohammadi 2005). This soil-borne pathogen (Wingfield ) infects host plants through mechanical injury and growth cracks or during changes of moisture levels in the soil (Suleman ).

An initial study identified Thielaviopsis paradoxa as the causal agent of black scorch in Qatar (Abbas & Abdulla 2003). In 2013, T. punctulata was isolated from black scorch-infected farms for the first time in Qatar (Al-Naemi ). This fungus produces single ovate aleuroconidia with a smooth or slightly rough wall and hyaline to brown cylindrical phialoconidia in chains. T. punctulata differs from T. paradoxa in the means of aleuroconidial production. While T. punctulata produces single aleuroconidia from conidiophores, T. paradoxa forms chains of conidia (Abdullah ).

The aim of this work was to sequence the whole genome of a virulent isolate (CRDP1) of T. punctulata, in addition to de novo assembly and functional annotation. Information from the fungal genome sequencing and annotation will allow us to identify fungal genes that are involved in pathogenicity and toxin production. The generated knowledge will help in designing an effective genetic control for black scorch disease.

Qatar: Doha, isol. ex Phoenix dactylifers (date palm), May 2013, F.A. Al-Naemi (BPI 893173 – dried culture).

This Whole Genome Shotgun project has been deposited at DDBJ/EMBL/GenBank under the accession LAEV00000000. The version described in this paper is version LAEV01000000.

CRDP1, a virulent isolate of Thielaviopsis punctulata, was grown in potato dextrose broth, and DNA was isolated using the CTAB method (Webb & Knapp 1990). TruSeq paired-end shotgun libraries and Nextera-based mate-pair libraries were generated and sequenced on a HiSeq2500, with a read length of 150 bp. The raw sequences were trimmed using Trimmomatic (Bolger ); reads shorter than 40 bp were discarded. SPAdes (Bankevich ) was used to perform de novo assembly of the genome. CEGMA (Parra ) was used to identify ultra-conserved genes in T. punctulata. MAKER v. 2.31.7 (Cantarel ) ab initio gene prediction pipeline was used for genome annotation. The MAKER v. 2.31.7 (Cantarel ) ab initio gene prediction pipeline was used for genome annotation and included the prediction models of CEGMA (Parra ), SNAP (Johnson ), AUGUSTUS (Stanke ), GeneMark (Besemer & Borodovsky 2005), and RepeatMasker (Smit et al. 2010). For the first round through the MAKER pipeline, we used results from CEGMA and SNAP to train the MAKER gene prediction models. The set of proteins obtained from the first round were blasted against SwissProt and the best hits were chosen as the training set for the second round through the MAKER pipeline. For the second round through the MAKER pipeline, we first retrained SNAP and then ran AUGUSTUS with the SwissProt dataset; finally we used those two gene prediction models to rerun MAKER in order to refine its own gene prediction models. Subsequent iterations of the MAKER pipeline proceeded in the same fashion as before, i.e. results from the previous round of gene prediction models were used to retrain and refine the MAKER gene prediction models. We stopped at iteration four when no further improvement was observed. InterProScan v. 5.48 (Jones ) and BLASTP v. 2.2.28+ (Camacho ) against NCBI-RefSeq (release 69, 7 Jan. 2015) (Pruitt ) were both performed to functionally annotate the proteins. MUMmer (Kurtz ) and QUAST v. 2.3 (Gurevich ) were used to perform a comparative analysis of the T. punctulata draft assembly with the publicly-available genomes of three other species of Ceratocystis s. lat.: Huntiella moniliformis (van der Nest et al. 2014), C. fimbriata (Wilken ), and C. manginecans (van der Nest et al. 2014).

Two libraries of shotgun (480 bp av. size) and mate-pair (5–7 kb) were prepared and sequenced using HiSeq2500 for the whole genome sequencing with total raw reads of 23 946 967 bp and 41 972 008 bp, respectively. The resulting draft assembly was 28.1 Mb in length (75x sequence coverage), and all contigs with a length of > 199 bp were submitted to NCBI’s genome database. The draft genome comprised of 2379 scaffolds greater than 200 bp, with an N50 of 92 723 and an L50 of 87 scaffolds.

To assess the completeness of the T. punctulata genome, we ran CEGMA in the draft assembly of the T. punctulata genome and identified 244 out of 248 ultra-conserved genes in T. punctulata (98.38 % completeness). MAKER v. 2.31.7 predicted 5480 proteins using the ab initio gene prediction pipeline. InterProScan predicted functional annotations for 4841 proteins; with BLASTP reporting hits for 2749 proteins with significant sequence similarity (expect value less than 1.0x10-3). We parsed GO terms from the InteproScan report and produced a list of unique entries containing 1636 GO categories (The Gene Ontology Consortium 2000). Additionally, we used ReviGO (Supek ) to cluster these GO terms and identify major represented categories. These categories include ubiquitin-ubiquitin ligase activity, endopolyphosphatase activity, urate oxidase activity, cyanate hydratase activity, dynein binding, RNA ligase (ATP) activity, acetyl-CoA transporter activity and chitin binding. Of note, some of the annotated genes belong to toxin-related genes such as necrosis inducing protein (NPP1) and cerato-platanin reflecting a potential functional role of the fungal toxin(s) in inducing disease symptoms.

We performed comparative genomics between T. punctulata draft assembly and the publicly available genomes of three other species of Ceratocystis s. lat., using MUMmer (pairwise comparisons) and QUAST (four-way comparative analysis). The genomes were those of Ceratocystis fimbriata (NCBI WGS assembly APWK00000000.1) (Wilken ), C. manginecans (WGS assembly JJRZ00000000.1) (de Beer ), and Huntiella moliniformis (WGS assembly JMSH00000000.1) (de Beer ). Though alignment results indicate significant similarity to the other genomes, they also suggest these species have less than 75 % in common between T. punctulata (CRDP1 isolate) and the other Ceratocystis and Huntiella genome sequences. Initial results suggest that these species may be related, but their genomes have undergone many changes since they split from a common ancestor. Altogether, our analyses confirm the degree of relatedness among the four Ceratocystis and Huntiella species previously reported (Wilken , de Beer ).

The annotated genome of T. punctulata (CRDP1 isolate) will provide a basis for understanding fundamental biological mechanisms in this pathogen. This will also lead to further understanding of the molecular interactions between T. punctulata and date palm, and eventually help in designing effective genetic control measures for black scorch disease in date palm.

Authors: O. Radwan, F.A. Al-Naemi, G. Rendon, and C.J. Fields