naraorr ‘Transcriptomie Analysis of Prunus domestica Undergoing Hypersenstive Response fo Plum Pex Virus Infection Transcriptomic Analysis of Prunus domestica Undergoing Hypersensitive Response to Plum Pox Virus Infection Ptah: Jaw 24,2014» plo og/10197 thus one 0100477 Abstract Plum pox virus (PPV) infects Prunus trees around the globe, posing sorousfrlt production problems and causing sovere economic losses. One variety of Prunus domestic, named ‘Joj', develops a hypersensive response to viral infection. Here we Compared infected and non-nfected samples using next-generation RNA sequencing lo characterize the genelic complet of the vial population in infected samples and to deny genes involved In development othe resistance response. Analysis of vral reads from the infected samples allowed raconstruction ofa PPV-D consensus sequence. De nova reconstruction showed a ‘second viral isolate of the PPV-Rec strain, RNA-soq analysis of PPV-infected Jo trees Identiied 2,284 and 786 unigonos that ‘were signicanty up- of downregulated, respactivaly (false dicavery rate; FDR«0.01). Expression of genes associated with defense was generally enhanced, while expression of those related to photosynthesis was repressed. Of the foal of 3,020, Sifferentaly expressed unigenes, 154 were characterized as potential resistance genes, 10 of which were included inthe NBS-LAR ‘ype. GWven thelr possible role In plant defense, we selected 75 addivonal unigenes as candidates for futher stuey. The ‘combination of next-generation sequencing and a Prunus vary that develops a hypersensiive response to PPV infection provided an opportunity to study the factors involve inthis plant defense mechanism. Transcriptomic analysis presented an Bverview of the changes that occur curing PPV infection as a wnole, and identified candidates suitable for further ‘unctonal characterization Citation: Rodamilans 8, San Leén D, MuhibergerL, Candresse T, Neumiller M, Oliveros JC, etal (2014) Transcriptomic Analysis of Prunus domestica Undergoing Hypersensitve Response to Plum Pox Virus Infection. PLoS ONE 86): €100477. (010.137 sjournal pone.0100477 Editor: ALN. Rao, University of Calfomia, Riverside, United States of America Received: March 3, 2014; Accepted: May 25, 2014; Published: June 24, 2014 Copyright: © 2014 Rodamitans etal. This is an open-access article distiouted under the terms af the Creative Commons ‘AtnDution License, which permits urvesticted use, dstibuon, and reproduction in any medium, provided the ofginal author ‘and source are credte. Data Availabilty: The authors confirm that all data underying the findings are fuly available witout restriction. All data are Included within the manuscript, Funding: The authors acknowledge support of tne EU in the frame of the FP7 KBBE-204429 SharCo project. Research is also supported by grants 8102010-18641 from the Spanish MCI. The funders had no role in study design, data collection and ‘analysis, decision to publish, or preparation af the manuscript. ‘Competing inter The authors have declared that no competing interests exist, Introduction Plum pox vis (PPV) isthe causative agent of sharka, a serious disease that challenges stone-frult production worldwide [1]-13) PPV is a member of the Potyvirdae ‘amily, the largest group of plant viruses 4) (6A single-stranded postive RNA molecule of 0 kb forms is genome. At ts 5-end, tne RNAs linked to the viral genomecinked protein VPg, and the 3nd caries a polyA tal ‘The genome codes fora large polyprotein and a truncated frameshift product thal are processed by lee sell-oncoded proteases into atleast 11 protein (7.8. PPV is transmitted by various aphia species in & non-persstent manner [9] [10] Eight PPV strains have been identified based on thle biolagieal, serological and molecular properties ae ablo to Infect a wide hast range of Prunus species [*1l-{"4]. One stan, PPV-D, regularly infects domestic plum (Prunus domestica), causing variable symptoms in leaves and fut [15]. Only very few P domestica variates show a hypersensitive response (HR) to PPV infection, e.g. Ké-Hybrde, OrtxStan 34 and Jojo [16}-(18) they fisplay necrosis on leaves and bark as well as death of new top sprouts, which stops val propagaton. oj. a descendant of the parent cuvars‘Ortenauer’ and ‘Stanly’ isthe variety with the largest numberof PPY isolates analyzed and the largest number of replications; its HR is elicited by all PPV isolates tested (PPV D, M, Rec. EA and W strains) [17]. This makes ‘Jojo’ an atracive Candidate for study ofthe factors invalved in this type of resistance. In plants, proteins encoded by resistance genes (R genes) tigger HR through director indirect interaction with avirulence proteins, inating a cascade reaction within the cell The majorly of cloned R genes encode ncleotde binding siteJeucine-ich repeat proteins (NBS-LRR,) making this family one ofthe largest. mast variable gene familes in plants [19]. Several studies have focuses bn resistance gene analogs in Prunus species, te natural host of PPV [20], [21], out to our knowledge, none, addresses the ‘mechanisms of Prunus HR to PPV. hitptjournas sos orgplosonaarticlo 10.137 journal pone, 0100877 3naraorr ‘Transcriptomie Analysis of Prunus domestica Undergoing Hypersenstive Response fo Plum Pex Virus Infection Next-generation sequencing, refered to as RNA sequencing (RNA-seq), has proved to be a valuable lol for assessing gene ‘expression ferences across the entre transcriptome for a wide range of organisms (22), 23). Unlike microarrays, these types of fnalyses can be performed when a genome sequence is unavailable, thus providing information an the bology of nar-model ‘organisms [24] (26). RNA-seq has proved useful not only for analysis of endogenous genes vanscribed inthe plant, but also for viral gename reconstruction ana recognition; talows study ofthe diversity of the infecting vial population, which is relevant for Adaptation and survival (27)-{30). Here we used ths tecnnology to compare gene expression between PPVnfected Jojo’ trees at the beginning of an HR response and that of non-infected reas. We parlormed two sues, ane focused on Vial reconstruction and helerogenelty analyses, and the other on endagenous plant sequences. The former alowed recognition of an unanticipated isolate of PPV-Res during the course of tha infection, andthe laller permitted de novo reconstruction of th plant transcrplome and assessmant of gone expression changes possibly related to HR in Prunus. In silico and qPCR tests confirmed the qually and strength of the results. Materials and Methods In year 1, one-year-old Prunus cerasfora‘Myrobalan’ seedings and in viro-propagated P. domestica 'Wangenheims’(Weivi) plants were plantas nan insect-proof greenhouse, Half of te plans, for Use ae reastock, were inoculated by chip budding in February wit a PPV-D isolate present in the Baden region, Germany, using budstcks from PPV-D infected P domestica ‘Katinka troes. One year later (year 2), plants ware tested for PPV by DASI-ELISA. In mid-May of year 2, two dormant buds of P domestica “Jojo were chip-grafted onto each plant used as rootstock and began to grow after four weeks. At the ist visible symptoms ofthe hypersensitivity response, young Jojo shoots ware harvested and frozen immediately in lquidrilragen. Leaves fram 6 or 2 plans, respectively, wero used to propare Jojo-M+PPV and Joje-W+PPV samples. Young Jojo’ shoots growing on S PPY.tto rootsiocks ‘wore sampled simultaneously by the same procedure, Frozen tssue was crushes in iqudnitragen using @ martar and pestle, Total RNA was extracted following the protocol of Carrier et 8. [31] modified by Pichaut JP, Labonne G ane Dallot S (unpublished: private communication) as follaws’ + g issue was Fesuspended in 20 ml preheated buffer 1 [200 mi Tris-HCI pH 8.5, 50 mM EDTA, 500 mM NaCl, 0.5% SDS, 1% PVP40000, 20 mg Proteinase K, 1% f-morcaptoetnanal}folowed by incubation (55°C, 15 min). The sample was aliquoted in 1 ml feachons. Kc (5 M. 500 pl) was added to each tubo and incubated on ice (10 min), then centrifuged (10,000 g,4°C, 10 min). Buller 2 (1/3 guaniéium Chloride 7:8 M+2/3 96% ethanol, 450 iL) and 60 pL silca suspension were added to each supernatant and silca was pelleted by Centrifugation (1000 g, 2 min, room temperature). Pellets were washed twice wit 600 pL buffer 3 (1/8 160 mM KAC, 23 mM! Tris HCI pH 8.0, 0.1 mM EDTA'2!3 170 mL 96% ethano). Silica was resuspended in 200i H20 followea by centfugaton (1000 9, 2 min) Supematants were incubated overnight with 187 yl 8 MLICI to preciptate RNA. Ina final step, RNase-free DNAse veatment (Promega) was performed to remove possible residual DNA, Library constuction and RNA sequencing were performed by the Beijing Genomics Insitute (BGI-Shenzhen, Shenzhen, China). Bey from 30 ug total RNA, poly-A¢RNA was selected by dligo-dT chromatography, folowed by RNA fragmentation, Fragmented IRINA was converted into double-stranded cDNA using randan-hexamer primers fellowed by end repar, 3" end adenylation and Adapter ligation. CDNA Fragments were selected by agarose gel extraction and enriched by PCR ampliication. The Horary was loaded onto an lumina HiSeq 2000 instrument for par-end Sequencing. The average read length of 90 bp was generated as raw data, Pir to assembly, FastOC software [32] was used to obtain information about the qualty ofthe sequencing data, This information \was used forthe inital fiterng of sequences by the FASTX Toolkit [33], removing the adapters and cleaning low quay sequences {mean «20 or more than 20% nucleotides wih Q<15) ane sequences with unknown nucleotides which could affect later Pioinformatic analyses. Virus reconstruction using PPV-D EF569214.1 as reference was done by mapping the sequences rom the RNA-Seq samples using the Bowtie aligner [34]. Selection of the most frequent rucectde fom the resulting alignments generated the consensus sequence, Ifthe numer of aligned reads at a positon was Zero, the rucleatide was designated 'N’.A secone alignment of the vral reads using the reconsincted viral enome as reference was carried out to abtain the final consensus sequence. De novo reconstruction ‘was performed by Oases assembler and the output transcripts wore searched agains! the val genomes ofthe NCBI nucleotide database (35) using blastn and blastx. Clustering using CAPS (36) ané CO-HIT (37] followed by extension ofthe genomic sequence was performed. As in the previous case of viral reconstruction using reference data, the de novo rebull sequence was used as Feference to map the val reads again using Bawte algner. ‘The population of viral sequences identied in the virus assembly was fitered to avoid possible PCR and sequencing errors, For viral genome algemant, oly pared sequences wera selected. To avoid ineertion and delelin errors, border regions af the sequences were discarded, and only 60 nucleotides ofthe central region of each sequence was considered. The frequency of each site-spectc helerageneity was calculated as tha percentage of mismatches to tne consensus sequence of the aligned reads. Since some hetoroganottios wore artifacts duo tothe sequencing process, the qualiy Score of each nucleotide road was used to computa the average probably of sequencing errr. Assuming that sequencing errors were Independen, when the numberof alignment mismatches for a cortain heterogenatt point was smallor than the expected numberof errors (the mean of te binomial dstnbution Bp) where pis the average probably of sequencing error in te position | and nis the coverage in tne positon i), that pont was considered a sequencing eror and removed from the analysis. entropies of the viral populations were calculated as descrbed by Wright et al. (38) hitpiournas.pos.orgplesonetarticle7d= 10.137 sournalpore.0100477 mnnaraorr ‘Transcriptomie Analysis of Prunus domestica Undergoing Hypersenstive Response fo Plum Pex Virus Infection For the analysis ofthe vial intraspecies, a previous quasispecies formation was done with QuasiRecomd software [39] using ruleotides 882-9588 duo to software coverage requirements, Recombination was not alowed to decrease complexity ofthe ‘nals, The ratio of non-synonymous-lo-synanymous nuclestde-substtuion rales (QN/dS) was assessed by Model selection program implemented in Hypry package [40] and employed for both intra-and interspecies studies. The SLAC [41], a maximum Fkelynood method, was used to identiy the global synonymaus nucleatige-subettution rate (aosteriar probablies bigger than 80%). To star the analyses, full genome sequences of PPV-D and PPV-Rec isolates stored inthe SharCo database [42] wore collected. Sequences were aligned Using ClustalW [43]. Consensus sequences as wel as nucleotide variation data were obtained trom the muttole sequence alignment. Transcriptome assembly was done with Velut [44] and Oases programs [45]. Input in both cases was a combination ofthe four samples and the best output, Velvet + Oases (mul-k), wae selected based on the NSO value, Subsequent analyses used the Contigs of is assembler. Oases was operated with mul-k option (K=47, 51,81, 85, 67,71, 75) and the option of minimum length of transcript was assigned to 200. Sequences were mapped tothe transcripts wih BWA 46) and distribution of the expression in the predicted unigere alignments was determined with RSEM [47] Diferential expression analysis was calculated with eogeR 4} Results were veualized wih the FIESTA program [49]. Toad functional information tothe genes, blastn and blaslx were employed Using several databases (NCBI EST [50], NCBI nucleotide [35], NCBI protein (51) and PRGab [52), and inlerproscan to determine Significant domains (e-value<10™ in ll cases). las(ZGO [53] was used to jin al ‘unctonal annotations and Io aban the enriched GO-terms associated tothe ciflarentialy expressed unigenes (FDR0.005) Peano nalyne ‘We used ClustalW for multiple sequence alignment 43}, flowed by searching ofthe bes fited evolutionary model implemented in 'MEGAS [54] with Protest (55). A phylogenetic ree was bul using the maximum fkelhood algoritum with tre parameters provided by the best protein model found by MEGAS [54]. The phylogeny test was done by the bootsizap methad. One thousand replicas ‘wore used to abiain the percentages a replicate trees in which the associated taxa clusteres (56, The first cDNA strand was generated fom 1 ug total RNA used inthe RNA-Seq experiment inthe presence of oligo(ST) 2.18 and Superscript Il reverse transcriptase (Invitrogen). PCR was performed in tiplicates using Fast Universal SYBR Green Mastermix (ROX; Roche) in an ABI7300 Real Time PCR System (Aoplies Blasysteme). Prmers were designed using the program Primer’ [67] as shown in Table S§_Gene expression was normalized tothe unigenes that matched TEF2 and RPI genes of P persica, Data ‘ere calculated by tne AAC? methad described by Pfafl [58] and are snown as tne x-old change in gene expression relative tothe control sample Results and Discussion In this transcriptome analysis, we analyzed four samples by next-generation sequencing. Dus to dificult in direct infection of “doja tees, PPV was inaculated by grafing onto infected roalstocks, Two samples, Jojo-W+PPV and Jojo-M+PPV, carrespand to RNA solated from Joo grafted onto infected Wangenheims (Welwa) ang Myrosalan (Myro) rotstacks, respectively. As control, we Used two experimental replicas, Jojo:Wt and Jojo-WN2, corasgonding to RNA from ‘Jojo issue grated onto nominfected Wein rootstocks. Data from RNA-seq ofthese four samples wore used for de novo assembly of contigs using Velvet [44], followed by (ases [45]. general summary of the dala oblaied is preserted in Table 1 “able 1. Summary of de novo sequence ansombly using Velvet and Osea hitplex doi org/10-137 foul pore. 0100477 :00% ‘Analysis ofthe Vial roads showed notable dferences between the to infected samples, Jojo-N'+PPV had 190 times fewer PPV reads than Jojo-W+PPV (2,203 and 330,438 reads, respectively), probably caused by experimental factors such as a stronger Tesponse to PPY by “Joie plants grafted onto Myro, cferent viral propagation rates between the grafted trees, or other factors Aiffcult to adress due fo the variable nature of tese infection experiments. A few vial reads were found during analysis ofthe non-infected tissue samples (80 reads in Joje-W1 and 132 in Joje-W2), possibly due to contamination during sample preparation or sequencing, bt should ot affect subsequent analyses (NCBI-SRA study accession: SRPO41925) Roolstacks used to infect “Jojo trees were previously inoculated using budsticks ftom PPV-D infected P. domestica ‘Katinka’ toes, The sequence ofthis PPV-D isolate was unknown, For viral reconstruction, a vial template (EF589214) was chosen by searching PPV-D inthe NCBI database [35] Alignment of the Jojo-W+PPY reads agains thi template allowed reconstruction ofthe neat- complete val genome, using the most abundant nacleotie at each positon to buld the consensus sequence (Figure 1A). PPV-D Isolate FR-B5pI rom the SharCo database [42] was the most siilar ral homologue of PPV ojo-W. hitpiournas.pos.orgplesonetarticle7d= 10.137 sournalpore.0100477 ans.naraorr ‘Transcriptomie Analysis of Prunus domestica Undergoing Hypersenstive Response fo Plum Pex Virus Infection ‘Figure Viral construction using rads fromthe JojosW+PPV samp. (A) Coverage ofthe vial genome was obtained by the template-based approach, Protein-coding sequences and untranslated Fegjions are labeled and colored. Nucleotide postions at ther borders ara marked. (B) Heterogenely analysis. Tho vial ‘genome is depicted in a colored bar as in (A). Line height indicates the heterogenety values ofthe Jojo-W+PPV population Felative tothe consensus sequence as determined by tne template-based approach, Changes that alter the amino acid ‘sequence are in red; region with no signicant heterogenelies i indicated by a black line beneath the genome. (C) PPV-D ‘ang PPV-Rec sequences obtained after speci clustering of the vial reads. Their corresponding coverages are depicted in ‘grey. Both stains are shown schematically, PPV-D genome (be), PPV-Rec genome (orange and white). Orange indicates ‘he part of PPV-Rec similar fo PPV-D; white indicates the part similar to PPV. Nucleotide change distribution between the ‘consensus sequences (grey) is shown in the space between the genomes, hitpex doi org/10.137 Vjoumal pore.0100477 4001 For heterogencly analysis, a postion was defined as heterogeneous when occupied by a nucleotide diferent from that ofthe consensus sequence. Changes were itered to avoid base riscallng in the sequencing or amplification processes. Atotal of 1,261 hucteotide changes were found, and the overal pattern revealed two pecular features (Figure 1B), the unusually high percentages of heterogeneties in most genomic regions, and the existence ofa region at the N-terminal pat ofthe capsia protein (CP) with no marked heterogeneties; ths was more string considering that this region is described as a highly variable genomic region in Potyuridoe (3) To test fora possible mistake in the viral reconstruction, we folowed a de novo approach thal also used the assembled contigs from infected samples that matched PPV sequences ofthe NCBI database [35], fel analyse revieved a consensus sequence identical fo that obtained using £*560214 as template. Detaled analysis of he resuts showed the formation of two minor contigs discarded in the fist assembly process, which belonged to the recombinant PPV-Rec svain. In ths sain, the genome from the 5 region to the Nib C-terminal region derives from PPV-D, whereas the remainder derives from PPV-M. The two contigs discarded during inital ceconstrucion belonged to the PPV-M part of PPV-Rac. Sk contigs from the PPV-D part of PPV-Rec wero sufficiently Sinilar to PPV-D to be included inthe frst sequence reconstruction, thus contributing tothe high genome heterogenelty, In contrast, heterogeneity ofthe CP N-terminal region, covered by reads from a single isolate, was correspondingly low. Once the presence of a mixed infection was identified, reads corresponding o one or the ther isolate wore distributed to reconstruct them and reassess coverage. There were 200,034 reads forthe PPY-D isolate and 97,188 reads that matched the PPV-Roc isolate; 17,832 reads could nol be assigned to one or the othor and were removed from Sunsoquent analyses. Now Consensus sequences ware obtained by d@ nove assembly, as previously detaled. The PPY-D Isolate consensus sequence didnot vary. The consensus of the PPV-Rec isolate had closes! homology tothe Slovakian isolato SK-1002ap in the SharCo database [42] (gure 10). The sequence ofthe virus that infected the Jolo-M sample could only be obtained using the template-based approach, because of the low number of reads avalable. total of 25 reads for PPV-Rec were Idontfed froma total of 2.154, this was a lower rao than inthe previous case, but not suprising given the variability of he experiment. The assembled PPV-O sequence had 99.3% Sequence similarly at the nucleotde level compared with the PPV-D in the Jojo-W sample (Figure S1A) There were nol enough reads fo assemble the PPV-Rec sequence in the Jojo-M sample. To analyze the diversity within the vial populations that infected ‘Jojo, we studied sequence heterogeneities. Changes were fiteres ta avoid base miscaling inthe sequencing or ampiication processes. Changes in both viral sequences in the Jojo-W sample were scallored throughout the viral genome, bu increased a the 3' regan (Figure S2A\, coinciding with sequance coverage. The entropy ofthe two populations, computed forthe validated sites, showed similar values (PPV.0=0.0040; PPV-Rec=0,0038), For PPV-D, we Fecorded a loa of 1,018 positans showing sequence heleroganeity, which comprised ~10% of the tal PPV ganome. Of these ‘anges, 655 were non-synanymous, In PPY-Rec, we found half the numberof total heterogeneities (584), butte 380 non synonymous changes mainained a similar nan-synonymaus-o-synonymaus rao (Figure 2A). There was no nelabie overlap between the heterogeneites ofthe two viruses infecting ‘Joo’ and coincidental changes did not favor non-synonymous selection, ‘These rests poin outa stochastic orgin ofthe changes, Since chip-budding infection eulas out the genetic dri sec of te inoculation procoss it can be assumod that stochastic changes are probably favored by the botloneck ofactimposad on wral populations by thei invasion of new leaves [59-(62]. Nonetheless, as the dNIGS analysis shows, a sight purfying selection can be onsidared in both casos (PPV.D=0.67, PPV-Roc=0 80) and this isin agroamont with results from a provious exporimont In P {lomestica infected with PPV-D, PPV-Rec and PPV, in which viruses deplayed week negative selective pressures [63], The idea of viral seclusion (58), 64] cannot be addressed here, since the sample analyzed Isa pooled sample from diferent leaves and trees. hitpiournas.pos.orgplesonetarticle7d= 10.137 sournalpore.0100477 ansnaraorr ‘Transcriptomie Analysis of Prunus domestica Undergoing Hypersenstive Response fo Plum Pex Virus Infection Figure 2. Vr heterogonehies analysis. (8) Heterogenotis in te two major variants in the Jojo-W+PPV veal population, PPY-D (top), with amino acid changes In blue; PPVRee (bottom), with amino acid changes in orange. (B) Varabily percentages of PPV-D (top) and PPV-Rec isolates (bottom) in the SharCo database [42] that afer rom consensus at each polymorphic ste, Amino acid changes in PPV-D, Due, and in PPV-Reo, orange. Coincidental changes in (A) and (B) are marked with grey symbols hitpsex dol org/10.137 Vjoumal pone.0100477 002 Sequences of full PPV-D (84 sequences) and PPV-Rec (8 sequences) isolates in the SharCo database [47] were used to buld two Consensus sequences and addrass infer-solato varabilly. In both cases, changes wore evenly distributed throughout the gonome (gure 528), PPV.D and PPV-Rec shawed 109 common variable sites ang, in contrast tothe Inra-solate heterogeneity ofthe “Jojo samples, changes at nucleotide twee of codon tiplels were more frequent than inthe ol two positions (27, 34 and 169 for fist, second and thi nucleotides, respectively) This s expected for an analysis of nte-isolate varablty performed on consensus sequences that have undergone evolutonary pressure. Wie found few high percentage heterogeneities (25%) in viral sequences inthe Joo-W; there were four in PPV-D, three of which were non-synanymous, ane fve in PPV-Rec (two non-synonymous). None ofthe changes were coincidental between srains. The hhan-coding change of ‘Joo PPV-D (position 2340) appeared in the SharCo PPV-D variabilly, and two nan-coding changes of ‘ojo PPV-Rec (positions 6597 and 9721) were also present in the PPV-Rec variability in SharCo. None of the amino acid changes in he “Jojo wiral soquences was found in the SharCo reconstructed consensus sequences (Figure 2B). Given the small number of related changes, ti diffcult to address thee staisal significance, but the fact tal they appear at the nucleate level supports @ random tigi. We analyzed the heterogeneities in the Jojo-M+PPV sample for the PPV-D reconstructed sequence as forthe Jojo-W+PPV. Distribution of cnanges was not homogeneous, and heterogeneties were again more numerous a the 3' end ofthe genome (Figure SS1B). There was no notable overlap between changes in this sequence and those inthe other infected sample. The pattern was nonetheless similar to that observed forthe inter-isolate analyses, with changes inthe thid nucleotide position favored over Changes in the other two (6, and 82 for frst, cond and thre nucleotides, respectively). ‘The combined reads from al four samples were used to assemble 117,919 transcripts that corresponded to 35,339 unigenes (Table 4), Blast search ofthese unigones was performed against four databases, Plant Resistance Genes database (PRGdb) [52], Prunus persica EST (50, plants EST [50] and rrint NCBI (35), (51). To analyze the number of mapped reads against the unigene Sequences, the BWA 48) aligner was used with defauit parameters. Statistical analysis was performed using EdgeF (48), a bioconcuctor package (www_bioconductor og) for RNA-seq. Total data were ‘tered to dently the dferetially expressed unigenes with statistical significance, using a threshold of false ciscovery rate (FDR) equal too less than 0.01. visual summary ofthe, results obtained by comparing te four databases can be seen In Figure 3A. A total of 3,547 unigenes did nat match ary sequence inthe databases used Figure 3: ferential expression of the oj’ sample unigenes {A) Venn diagrams crossing allunigenes in the four databases as indicated (tp) and crossing diferentialy expressed Unigenes in the four databases as indicated (potom).(B) The smear plot represents the taguise log FC (fold changes of tnigene reads in infected versus non-infected samples) against log of average number of unigone reads inthe sample set per milion total reads (RPM) of all unigenes. Green and red dots represent up- and downregulated unigenes, respectively, ‘with an DRS0.01 hip dot. org/10.137 journal pone.0100477,g003 hitptjournas sos orgplosonaarticlo 10.137 journal pone, 0100877 513naraorr ‘Transcriptomie Analysis of Prunus domestica Undergoing Hypersenstive Response fo Plum Pex Virus Infection ‘The 36,839 unigenos are dopicted on a smear plot (Figure 2B). toll of 3,020 unigenes were found tobe iffrentaly expressed in nfecied versus nominfected samples; 2.234 genes were upregulated and 786 genes were downregulated (Table S1). Homologs for 2,843 of these urigones wore detected inthe PRGdb, P. persica EST, plants £ST or mint NCBI dalanases. To describe the main pathways modifies in Jojo plants during PPV infection, we studied the ungenes by Gene Oniology assignment, of GO-term (hypergeometric test, FDR<0.006) (Figure 4) Cell proliferation and motor activity were highlighted among the enriched terms in biological process and molecular function categories, respectively. Genes associated with cellular components such as extracellular region, cel wall, cytoskeleton or chromosome were upregulated; photosynthetic electron anspor, electron carrier activity and {thylakoid were downregulated terms ofthe three categories (Figure 4). Analyses ofthe unigene coding proteins by protein domain search using the Interro database [65] retrieved groups of proisins similar to those oblaines by GO-term search. Thus, proteins involved in chromosome maintenance, motor and kinesin activites, nydrolases or cyclin proteins were among those Upregulated:gltaredoxr-lia proteins and cytochrome P50 were downregulated. Interpro alowed detection of proteins with other domains tha might be relaed to defense response such as baumatine (66), paces [67] or proteins with leucine~ich repeats (gure 4) Figur 4. Gane Ontology (00) enrichment analysis of up-and downregulated genes Bie boxes separate the ferent areas of GO-Term distouion. The purple box deliits the Interpro domain distabuton ‘Values are inversely proportonal tothe False Discovery Rate (FOR) as calculated by Blast2GO. Bar plots corresponding to Interpr functional domains are simpified,jining related functions, following the Interpro relationships inbertance data, hitpex doi org/10.137 Vjoumalpore.0100477 3004 ‘A ganeral examination of related studios [68}-(72] underscores two shared fealures, repression of photosynthetic function in neerotie issue and enhancement of defense gene expression, these common features are also observed in the present analysis. A Search for more specific terms, such as cel wall or metabolism, reneved contrasting resus, which is unsurprising qwven the broad differences between studies. The overall view presented in our study resembles those shown previously and isthe general idea of 2 call experiencing pathagenic stress and undergoing profound changes, with modifications in protein activty, cascade pathways, trafficking ard morphology. eho onary expressed vias ne arte PRGA eve To determine more directly the laments involved inthe hypersensitive reaction, we studied the genes implicated in plant defense and resistance. Of the total number of unigenos, 1,084 matched entries in the PRGAb, of those, 754 wore expressed diforontaly after PPV infection, 126 up- and 28 downregulated (Table S2) More than 50% ofthese unigenes corresponded to proteins with suspected kinase activity, and 13%, mainly among those upregulated, coded for LRR receptors. Moreover, a small percentage of resiatance-related upregulated unigenes (4%) encoded prateins with presumed nucleotide binding activity. Overal, -20% of the Unigenes that matcned PRGdb sequences showed miscellaneous functons and 10% of he total had no defined function. Due to the importance of NBS-LRR genes in esistance to vial infections [73], we focused specifically on these types of genes. In general, ater pathogen detection, F genes unleash a cascade mecharism win te cel that results in HR and systamic acquired Fesistance signaling. For the planito provide an organized reaction, R gone expression must be tghly controled dung patnogen attack; nonetheless, the mechan'sms responsible for this regulation remain a mystery. There are reports of R gene upregulation following infection, indicating a dosage effect [74]-(76], whereas others describe R gones for which expression levels remained unchanged during HR, suggesting a cistinct activation mechanism (77), (78). In our RNA-seq analysis, we identified 314 unigeres as members of the NES-LRR family, of which seven were upregulated. Sze and dstsbution ofthe reconstrucied unigenes fs shown in Figure SA, Four of the upregulated unigenes appeared to code for Tobacco mosaic vnus-resitance N-ike proteins, witha Tol intereukin-lke receptor atthe N-terminal pat (TI-NS-LRR). The other three encoded proteins with a col-colled region atthe N terminus (CC-NBS-LRR), with ro specie gene sila. Figure 5. NBS-LRRype "Jojo unigene ieretlly expressed as ares of PPV Infection. {A) Scheme showing unigenes defined as C-type (red) or TIR-ype (blue) according to their sequences orto those of closest Felaives identified by Blast analysis. Fold changes (FC) are depicted above each unigens. (B) Phylogenetic ree using the \vanslated sequence of Jojo unigene 452 and the closest homologues in P. persica (80% or higher) of sx ojo’ unigenes. NBS-LRR Wanslated genes ftom F vesca, P trichocarpa, V. vinifera, M. domestica, and A. thalana were selected from the hitptjournas sos orgplosonaarticlo 10.137 journal pone, 0100877 ensnaraorr ‘Transcriptomie Analysis of Prunus domestica Undergoing Hypersenstive Response fo Plum Pex Virus Infection NCBI database based on sequence similarity. The N protein of N. gluinosa was also included inthe ree. Color code as in (A) withthe adation of green forthe outsider protein used to root the tree. Curated R proteins are marked next tothe gi number. Boolstrapsing numbers are located nex! to ach branch. itp doi. org/10.137 Vfoumal pore.0100477 005 Three NBS-LRR-rolated unigenes were downregulated (Figure 5A). One of these belongs tothe TIR-NBS-LRR femiy, whereas the other two are ofthe CC-NBS-LRR type. To our knowledge, there are no reports in which R genes deseabed to be involved in HR. linderwent downregulation during infection. The closest scenario was deserived for Arabidopsis thaliana plans in which the RPM protein, which confers resistance to Pseudomonas syringae, was degraded as HR progressed {73}. This downregulation occurs at {the protein level, but the fnal results analogous to a decrease in mRNA accumulation, One ofthe CC-NBS-LRR unigenes shown hera tobe downregulated, unigene 22.213, resembles the RPMT gene, suggesting a similar mode of action. Six of the diferentially expressed unigenes can be assigned to homologous P persica gones (>80% identiy) and ther translated proteins can be used to buld a phylogenetic ree wih their closest NBS-LRR relatives from other plant species whose genome has boon completly sequenced, Fragaria vosca and Malus domestica wore gelectes as other Rosaceae species, Populus tichocarpa and Vis vinifera as representatives of woody plants, and A, thalana as the mostused model plant species (Figure 5). Since the Complete unigone 452 Vanseript was reconstricteg, bath its \ranslated protein ard ils P. porsica homolog wore included in Ure Phylogenatic analysis. The N protsin from Nicotiana gutinosa was also included, and a resistance protein from the PRGdb that cid hot bolong to the NBS-LRR type was used as an outgroup fo root the treo. In all cases, tho F vosca R protein was the closest Felative tothe translated P.persice-elated unigenes, and the fopology ofthe tree suggested that they were bona fide homologues, The t29 showed separate clusters of N- and RPM-ko protons, incliding the corresponding Prunus hornologues, which suggests high conservation of these proteins. In the other cases, the R protein homologues identified did not show a defined functon, ceva PRG be nce ese ‘Among the 3,020 unigenes dilferealy expressed during PPV infection, some wore not describad as R genes, bul might bo involved in plant defense and HR. To perform a detalled gene-te-gene analysis based on iterature ndings, we frst narrowed the Unigene population. The unigenes selected wera hase absant from the PRGab an with an absolute FO>20 and an FOR0.0% This reticed the Isto 703 unigenes, @ manageable number fora comprehensive study; ofthese, 75 were chosen as possibly involved in plant response to PPV (Table $3) The list included a wide range of genes coding for proteins as diverse as Pistones {80}, aspartic proteases [61] or polyphenol oxidases [82] and define a clear starting point for future research on plant defense, Read quality was assessed fllowing well-established parameters in RNA-seq protocols (gee Methods). Data were futher validated by the statisical parameters ofthe programs used. As an additonal checkpoint, we performed a so-called fld change tes Reconstructed unigenes do not necossally cover the entire eoquence of a gene, and two oF more unigenos can correspond tothe ‘same gene; these sequences would present similar up-or downregulation expression pattems. For simplicity, tis test was carried Cut only with te diflerentaly expressed uniganes in PRGdo (52) In this database, seven genes wore matched by more than one Unigene, and fold changes were consistent inal cases (Table SA). To illustrate this test, two reprasertative examples in which Several unigenes wore part ofa single gone are detailed graphically (Figure S3). trons, Five unigenes with distinct expression pattems wore selected for confirmatory analysis by qPCR. Due tothe lack of data forthe domest plum, two reference genes were selected, TEF2 and RPI, based on Ine work of Tong etal. [83] in peach trees. Average ‘and individual results are shown in Table 2, together with the resus for these fve genes in RNA-seq analysis. The remarkable level Of reproducily ofthe data obtained using these two approaches validates the consistancy ofthis tanseriplomic study. ‘able 2. omparion of PCR and RNAea dat hitpex dot org/10-137'Vjoumal pone.0100477 :002 Conclusions. RNA-seq has praven tobe a valuable technique to study gene expression profs in madel and nor-madel organisms. To aut knowledge, this iste fst report that attempts expression characterization of Prunus trees developing an HR response to PPV infection. This sty alowed nat only reconstruction af the PPV-D viral genome consensus sequence, But alsa analysis ofthe genetic complexity of the viral population, which led to identifcaton ofa second, previously undetected major val Sain, PPV-Rec. ‘We identified 3,020 unigenes that are ctferenally expressed during infection, 164 of which are rlated to genes implicated in defense responses. OF these unigenes, 10 wore of the NBS-LRR type. Closer analysis of the data dented addtional canclidate genes that show lage variations in ther mRNA levels and might be Involved in vial resistance, establishing the bass for further Tunetional analyses, ‘Supporting Information Figue st Viral reconstruction using reads found in the Jojo-M+PPV sample. (A) Coverage ofthe viral genome, Protein-cocing sequences are labeled and colored. The nucleotide positions al ther borders are marked. (8) Heterogeneity analyss. Tho viral genome is depicted using a colored bar as in (A). Line height indicates the heterogenelty values ofthe Joj-M PPV population hitpiournas.pos.orgplesonetarticle7d= 10.137 sournalpore.0100477 msnaraorr ‘Transcriptomie Analysis of Prunus domestica Undergoing Hypersenstive Response fo Plum Pex Virus Infection (above the genome) inthe indicated postion, Amine acd diferences are depicted in red. Below the genome, grey and red nes Fepresent silent ang non-silentaifeences,respecively,relave to Jojo"W*PPV consensus Sequence, The green line shows an amine acd change in PIPO. 4oi:0.137 Houma pone.0100477 5001 (rr) Intra- and inter isolate heterogeneities analysis. (A) Intra-solate heterogeneities distrbution in PPV-D (blue) and PPV-Rec (orange) from the Joje-W+PPV sample. Heterogenetis are grouped by 490 nucleotides. Grey nes between fe two stains mark changes common fo both, (B)Inler-solate variably dstnbution in PPV-D (Blue) and PPV-Ree (orange) from the SharCo database [42]. Polymorphic sites are grouped by 490 nucleotdes. Grey Ines between the two strains mark cnanges common to both 49110137 ijournal pone 0700477 5002 cr) Parallel fold change test. Schemes of two genes from the PRGdb [52] that are matched by several diferentialy expressed ‘Jojo’ Unigenes. Greon (Wi and W2) ane red (W and M) boxes show the numbers of reads corresponding to non-infected and Infected samples, respectively ai-10.137 sjoumal pone.0100477 5003 TF) Differentially expressed unigenes. The raw number of reads found on each sample is given in calurmns Jojo-W,Jojo-W2, Jojo. WPPV and Joje-Ms PPV. Fold changes between infected and nornfected samples and FOR values are detailed in colurnns FC land FOR, respectively. Column labeled Description expands on the details ofeach unigene, including length in nucleotides. 4oi:10.137 ijoumal pone.0100477 5004 bas) Differentially expressed unigenes found in PRGdb. The raw number of reads found on each sample is given in columns Jojo WI, Joje-WW2, Jojo-W4PPV and Jojo-M¥PPV, Fold changes between infected and non-infected samples and FDR values aro detailec n columne FC and FOR, respectively, Column labeled Deseripion expands on the deals of each unigene, inlucing length in nucteotces, 4oi'10.137 ijouralpone.0100477 5005 os) Genes possibly involved in plant defense not included in the PRGdb. Similarity toa known protein is given in the Homology column. The raw number of reads found on each sample 's given in columns Jojo, Jojo-W2, JojoaW/+PPV and Joje-M=PPV, Fold changas between infected and nor-infected samples and FOR values are detailed in colurmns FC. and FOR, respectively. Column labeled Deseriplon expands an the delale of each unigene, Including length in nucleoids. 4oi"10.137 ijounalpone.0700477 5008, bas) List of genes from PRGdb matched by more than one ‘Jojo’ unigene. Reference numbers ofthe matched genes are shown in the NCBI Matcn column. The raw numberof reads found on each sample is shown in columns Jojo-W,Joja-W2, Jojo-W=PPV and Joje-M=PPV. Fold changes between infected and non-infected samples fs given in colurmn FC. Column labeled Description ‘expands on the details of each gene. 4o1-10.137 Vjoumal pone.0100477 s007 os) Primers used in qPCR experiment. 4oi*10.137 iJoumal pone.0700477 5008, (0c) Acknowledgments |We are grateful o Beatriz Garcia for technical assistance, to Sylvie Dallot, Gérard Labonne and Jaar-Philipe Pichaut for providing ‘tho protocol for total nucle acid oxtracton, to Miguel Arenas for assistance inthe analysis of seloctive pressure in the viral popilaton, and to Catherine Mark for editorial assistance, ‘Author Contributions Conceived and designed the experiments: BR JAG. Performed the experiments: 8R.LM MN. Analyzed the data: BR DSLTC JCO. JAG. Contributed reagents/matenals/analysis tools. MN JCO. Wrote tne paper. BR DSL TC MIN JCO JAG. References 1. Camtra, Cate N, My A. lSca 2008) Pm poe vin a th stats ants asso wih sharks sate. EPO Bll 36 202-208 hitpiournas.pos.orgplesonetarticle7d= 10.137 sournalpore.0100477 ansnaraorr “Transcriptomic Analysis of Puss domestica Undergoing Hypersensitive Response fo Plum Pox Virus Infection iow Alo » PubMedNCB! + Gooel Scholar 2. Lipez soy Js, Femnex Feder MR, Canta M, Gaia JA 2000) Solel sepia slum px vus ltectna 76: 121-186, iow Aielo » PubMedNCBI > Gooel Schl 4. SocharJ, Baba Adam V Keke 8, Kish 2012] Shara: the ast, Ne rset ad the ae, Vines 42855-2005 Viow Alo » PubMedNCB! + Gooel Schl 4. Adan Mi, Zeb FM, Fone R Rabestin F, Senger DC. a al. (2012) Fary Payne. I A.M. King, M. JAdons, E,BCastens ant “eon esters. Vn Toray. Oxon Eloos pp. 069-100. 5. pes-MoyaJ, Gara JA 2008) Payvses. BM. Mary ans MH. Vn Regonmorl, eters. Eeyelapein of Viloy, Tin Eton. Oxf bev pp. 313922 6. Ivanov KI, Eka K, Lamu A Makinen K 2044) Maca ané Cellar Mechanisms Undying Payvins Iveco, J Gn Vil di to. toeve0 942200, View tila © PUBMedNCBI + Goole Schl 7. Gavia JA, lesa M, Cama M, CactosseT 2019 Pim por vin ad share: 8 mode oy and a maj oss, Ma Plant Pato! 15: 26-241, ‘ew Atle » PUBMedNGRI + Google Schl | S4trZ, Glas M 2012) Unoing the seca of pli gowns from eplemilogy to goramie, Acta Vel 6: 217228, ‘Yew tle» PuBMedNOBI » Google Schl 4. Glow FOamsteogV, Stns, Schneider, Lite a. 2008 Pin po in rath arc: neato of apis vectors anda patente or ‘hatin vis spend. Pryosthony 86-874. iow tele PubMedNCBl = Congle Scho 10 Labone 6, Yoon M, Cut JB, Avira Lee (1865) Asides pte vectors of Pen pox vn: Corpus of melds of etn art pdemlaalconsoqan:es. Acta Hee 386: 207-218 ‘iow ile = PubMedNoB! + Google Schl ‘1. ancsse 7, Gaara (06) Causa gent of sharks dsoase:histrcal perspective and cuentas of lim pox vis sirans. EPPO Bull 3: 289 iow tle » PubMedNCB! = Google Schl 12. Glau, Pinos ¥, Prada L, Nogjova A, Shraydr al (2018) Chrctrzation of serene sol of Pum go vn Fa te Veg bia Russia vene anew chary sano he ves, Pytpateogy 10% 872-978 View tle» PubbedNeBl » Goal Schl 1% Leer 6, Corora M2006) Hoss a rpms ef Pum pox vis futing Pras spaces EPPO Bull: 218-27, View ile © PubMedNCBI » Google Schl “1. Uubag Serge, CantessoT Svandls Duna rts L, Gz Mea. (2008) Further characttzaton ofa row reamtbnant group of Pl pox ius les, PPT, unin excars the rar proviso They. Via Ras 12121125 iow tele = PubNedNCBI = Google Scho 15, Fuche M, Cx R Ca 2006 Plan pox enese flan IPM Fact Shee. vali: p:ysiomcomal edutetsneststreetsoasespoep pt Accessed 2014 Jun 2 16 Uertnsoge, Near, TaatrO, Hertrann W (2010) The near fhe nyprrnsy relstens ot Expean pin (Prim domes.) paella poe vie Jace Kobe £27 527-328 iow aria» PtbtedNCB! = Coot Schl 17 NaurierM, Hartman W 2006) Tha prea quatative nse of hyparseneiily of Cuopetn pum (rus comes agaist he Pen ar ts ents deserpin us he hypersnscvty nex. Hee Se 0-54, ‘iew Mile » PuMedNoBI + Google Schl ‘18, Nounitor, Harman W,Stsse (200) The hyprsrstvy of Euopsan pm eganst Plum por vn (PPV) @ prmsing mechanism ot ‘matte, Pylopacloga Poona 7-82 iow arelo » Pibted NCB! = Geos Scholar 18. Mote P (2008) Mechanic rcopiton in inant gan ede reitnee, Ay Vi Rs 781-28, View Alo » PubtedNCBI » Geotl Schl 21, DecoargV, Feulrgre W, Lamar? Cll OL Marin, sa. (2005) Anagues avin eine gees map to GTLS far istarce ashes ‘deearen Pras aavaans Hol Geet Gnomes 272 620-6, View Mtcle » PubMedNoB! = Googe Scholar Soraro J, Viarova 8, Romere Cicer G, Baden ML (205) Chractenation ans meppng o NBS-RE fsstarce gon analogs in spot (rus ‘smonaes L)- Thea pl Gane 10 980-09, iow aticto » Publ NCBI > Gol Scholar hitptjournas gos orgplosonaarticle7id= 10,137 journal pone. 0100477 onsnaraorr hitptjournas sos orgplosonaarticlo ‘Transcriptomie Analysis of Prunus domestica Undergoing Hypersenstive Response fo Plum Pex Virus Infection 22. Warin JA Wing 2 @1) Nex-geneton ascii essen. Net Ry Gere 2: 671-682 Yen stele = Pubbed NOB! * Goole Schl 23. Vong, arsenM, Sry M (2008) ANASex a alone or tances. Nak Rv Gana 1 67-83 View ile PubledNCBI = Geol scholar 24, Gavin 2 Singh HR, Shankar R, ShamaN, Kura A, al 2012) Os nove sequencing an chacletn of Perirn kur nace a temperatures showed major tancelome estos, MC Geran 1.128, iow Atle = PuslaaNCB! = Google Schl 25. Kanahare- Mi R, Wide K, Aci N. Chie (2011) Expression pting wine parame saqunes naman a oneodel pei, Pash shin (Padus vot) by nex-genetn sosuaring.PLaS One 288. ‘iow ile» Pubted NCBI = Gotle Schlr 28. Xu DL, Long Uarg J, Zhar J Chan X, ea. 2012 De aves terzsion of ha ret kansortome of Aegioes varie sng an ateracion wih be cel eat Aenalae, BHC Ganames “3°58 iow Aetele = PubladNCB! Google Schl 27 Kreze JF, Pare A Utes M, QlepeO, Fucis 8, (2008) Compl via nuts sequace and eecovry cf vovel uses by deep ‘iow aie» PubledNCBI = Geotl Schots 28. AsO, Luo YR, LAL La EG, 6a 2010 Vi escovary by deo sensing and asorly of vine devs smalstencing RNAS. Proc Nat iow tele » PuBMadNCB! = Google Scholar 23. Zag 0, Dauner, Baia ,BewterinkalN (2012) Rea angth vrs death of everage fo vl eusepec ‘oH, Teng A Ln CH LoS, Fall P ot (2012) Garamie sequen ans compare sali of Epson ar vie namical om prenary aasphanyngelcatenona Dope. PLS One 7: 6390, 31 Carer, Sato 5, Roser-Coud M, Canaguer A, Kectho A, a 20M) A licen and rape poco eplantrele’ONA preparation sue fer Yew ale © PusledNOBI = Gotle Scholar 532. Bataham Birfomatcs.FASTQC, Aquat for hh hugh semuenc data. np awa orn babar. kinase 233, Harmen FAST Teo ip naar sh edits oceinden Hn Accesed 2014 Jun 2 1, Ssebag BL, Seven L201} Fat gapped lane with Bowie 2, Net Meads 8 257-59 iow Atle = PuBMedNCB! = Google Scho 235. NCBI cco ah tpl eileen gowmceee. Accessed 2014 June 2 28 Nang H, Madan A (1909) CAPS: AONA Sequace Aesembly Progam, Gane Res 9 856-87 iow Atle = PuniedNCBl = Gong Serle 37, FLL NUB, 2huZ, Wu. LW O12) COAT. scaeted fo clisteng the next-gen seauecing de, Sioratis 28: 350-8152 iow Atle» PubMedNCB| = Google Serb {8 wht CF, Mor Md, Teta G, Knoles N, Ht Pl 201) Beyond he consensus: asectng whines! val popu erly arasnaih deca vive by onngrex-qenrtin gorome seqionig, Vl 8 2282275, 2 Tel Zager, Prakaran 6, RotnV, Haga Et (299) Pobabliae eens of vel qasiepeies ste foresonsnaton, J Comput iow ale» PubbedNCB! + Gootl Scholar {48D W, Poon AF Frost $0, Kenskovsy Pa SL (2010) Dtankty 2010: site of phage andy ol for evasonry ily Bofemaies 26 2456-2487, onntovshy Pond SL, Fost 0 (0D) Nt so arnt ater al: a commparson af shade for scr amina a sts unser slectn, il Evol 10.137 journal pone, 0100877 rosnaraorr “Transcriptomic Analysis of Puss domestica Undergoing Hypersensitive Response fo Plum Pox Virus Infection 42, Pum Pax vin Ona, SHARED: SHARKA.CONTANMENTIN VIEW OF EU EXPANSION \hipv0plarataninn Ane, Accessed 2014 Jun 2, 45, Laten MA, Slckshelés 6, Bown N2, Choma R, MeGetigan PA. ot (207) Csi W ard Cust X vrsion 2° Blonfomatis 23 2887-258, Yiew tile» PURMedNCBI = Gon Scola 14, Zerino OR Bie (208) Veli grt for dene sno rea sembly ing in oh, Ganame Res 18: 821-875, ‘Yew tile». PubMedNCRI + Goole Schl 145, Sohle MH, Zein DR. VingonM, Bey 2012) Ose rob de novo RNA. assay actos the dynam rings a expression vel, iow Atle = PUBMeaNCB! ~ Google Schl 48, oubin 2010) Fat ant sceate ones grat with Surons-Wheel nstom: Saalamates 25. 585-585, iow tele PubMedNCBI = Google Schl 47-8, Deney ON 20") SEN: acute taser quatcton fom RNASeq dt who wile letane ganame, BMC Bama 12:32 iow ariclo » PuMeaNCBI = Google Schl 148, Ratingon MD, MeCar DJ, Smyth GK (20:0 edge a Bacon package for eet expression analysis fig gene expression de Bofomaes 26 199-40 View ile © PUOMGANCB! + Goole Scholar 4. olvoros JC 2007 FESTA@BIeinoGP.An interactive sever for amazing DNA miroaay experiments with plete. nipsbiiiogp cri eloaBiIESTA, Azceased 2014 June 2 St. NCB EST Mure Am gvucel, Accessed 2014 Juve 2. ‘1. NOI Pain (a tps tamaincinin nh govpctin, Accessed 2014 Jie 2. 52, SarcovarnaM, Rams G DeSimone, Fara Mato S, ta 201) PRGA: onlomratce plore par eitanca gune anys. Nile ‘lo es 36 Date. ‘iow Mile » PubMed/NCBI + Goole Scholar 13, Gonosa A, Gite 8, Garam JM, Trl J, Tan Ma (208 Blst200: a unvora aforanataton suakzatin ard ants ronal iow arto » PUBMedNCB! + Oost Scholar ‘54. Tenuta K Pelosan 0, Pete StocetG, Naa (2011) MEGAS: molecu tlio gna veer @starce, and maxrrum perseany msthods, Mal Bol Evl 2% 2781-2738 iow tle PUbMedNGRI + Gone Schl says ung maxima alton, 58, AbssclF.ZarsoyaR Pose O(n Proto ‘View Mile © PubMedNEBI + Googe Scholar setectn of bsttt models of rotate, Balfematies 2: 2104-2108 158 Flsnstin (186) Crtidnce Its on pylogores: An apoach us the bots, Evoliton 3%: 788-791 View Mile » PubMedNoBI + Google Scholar 157, Rozen 8, Seats H (2000 Peer on tha Yr general ses ans for Bolegst programmers. Methods Mol Bol 12: 35-296 ‘Yew Mile © PubMedNCRI = Google Schl 58. Prat 18 2001) A naw matrenatcel model orelavequarisien a relsime RE-POR, Nucla Acs Res 29 of ‘Yew tile « PubMedNOBI = Goole Schl ‘58, Jnd\© Marin JF ManeJesanaV,Lebnrs , Blane S006) Dane wal peptone aerate and woke ndepenerty = nal prema host an i 80: 2345-2567, iow Atte» PubtedNCBI + Goosl Scholar 1, Rosesinck Md (2004) Gant btlanecks teticepapultin vston an expatratal RNA vis peplaon. 4 Vel 78: 1056210587, iow antcio » PubledNCB! » Google Scr 61, Kate 8, Ered CA, Pear K (200) Mute hat tavers atft pli Waking mice, PLoS Pata 4 et00082 iow ale = PuBMedNCB! = Georl Scot (2, Savin 8, Mabica JM, Fre A Gacie-AreralF (2008 stimatin of eptonotlenecks dung estamie movement of abaco moss vn a tovecee ats. J Vio! 7? 8606-911, View tle». PUBMedNGBI + Goole Schl 163 Prada. Subr 2, Candas, laa Mt (202) Evasion of ne gna vray of Pl poe wns na hau oe, i Rs 187 182-117. ‘Yew tile » PUOMedNGB! + Goole Schl hitptjournas gos orgplosonaarticle7id= 10,137 journal pone. 0100477 ns.naraorr ‘Transcriptomie Analysis of Prunus domestica Undergoing Hypersenstive Response fo Plum Pex Virus Infection {4 ieireh c, Mais 200) Fareson Inn oven spat! separion of pawns sepUntons in mixed fot Netara peamara lan, J iow tele = PuBMedNCB! - Google Schl 65, hurt, Apa R wood TK, Boch A, Baran Ae (2008) ner. te tegen signe cabs, Nachle Aids Re 37 (64. Ki Ml, Ham BK, Kim HR Le), Kim Ys, a 205) nro ane in lara rteacon evidence between Net tabaci faut tin (TLP%) end eucaber mou vin pli. Prt Mal il 58961998 iow tile» PuabedNCB! = Coole Schl (67, Ko 81, JHC, Lae 8, Nam MH, Cusg JH. ls, 2008) SDSL tpaseKe 1 agalesoysaniceistance essere wih eter sigan a rmicopni Plant 8 25-285, ‘4. Batu, Gifts JS, Hong TS, Nang A (008 tered gene expression changes Arabidopsis et ees ard protoplats inesponse te Pl po iow Atte = PuBMedNCB! + Google Scho (8, Datck © 2007) Corgetve engresson song of Nein barthamine eves systray cla wih hve fi eves, Mal Par Mrabe pra 20° 004-1047 Yiew Mile © PUBMGANCR! + Goole Schl 178, Dizivanoas 9 Clonerte Moron My, Rubio M, Olmos E, Gala JA. 2908 Alero nthe eons metal lacs ROS ceualton pea ats in eaprse la lon pax ve. J Exe Sl $8 2147-260. iow tele» PubedNCBl = Gol Scho Pochtco R, Gaeia aves A, Marzane A de acaba MG, Conanes 6, a al. (2012) Congeratve aryl of taneiplomi an tamaalespnses to ‘somgale and neoratl lati iaracton tht leat el nth, Ma Plan Microbe Ietewat 9-70123, "72. RoargnG, Caray, Rur-Fore¥, del To Fy), lve, ea 2012) metn-anaysls evel th commanates ee rence in Aros Maton reagan ore vlpethagers. PLoS One 7. tS iow Atle» PuaMedNCB! = Google Schl 73. Guu MA Venton J, Ubdyaya CP, Nectar A, Pandey SK, ea, 2012) Plant dtcae eistace gone: Curt tts and Ue dats Pry Mot Pat Pata 8 1-8, 74. Levy M,Eeloau 0, Sala 2004 Tobacco masa vn regs the expression ofits oun restance gee N Plat Phys, 19: 29922987, 7% Mohr TJ, Maria NO, Ho 7, Weslo BU, Jlosko J, (2010) The Arabisopss downy mien estance gine REPS induced by pathogens an aati acd adi eles by W bx ls hme. Mo Pl cab Inara 25 1S05-1315, iow ate » PubMedNCBI = Goole Schl 78 Yoehinur 5, Yananaus U, Klay TiS, Wary 2X. a (896) Exeesson of Ket, e beta bighesitone nein ee, ence by Doctoral nolan, Proc Nel Acad Se: USA 95: 166634 7 Mos J, van Door AA, Wire J mors, Comelsson Bt (200) Expression of he Fusrom rsstance gene 12 caloalizes wih tho st of iow Atle = PuoMed NCB! ~ Google Schl 78 TenX, Mejor 8C, Koz A, Vist MA, Morgue Mt (2007) Gal expression aalysit of nucetde Hr sie lacie Hoh repeabencaing an 178. Bayes. Nam J, Dang J. 1998) The Arbigopss hala RPM ceeas rosstance gine pode! sa pepe plasma membrane pin thats ‘iow aril PubbedNGBl = Gol Schl aK, Fae CMa 201) Chranatin cartiation 8 bteald in placer nlaratos, Plat Phyl 157; 596-54. iow Atle » PubtedNCB! = Genel Sct 1. Maram Manzane MA, Tanaka, Obs OR, Yate RY (2010) Scluetuneton charateeabon fhe recombi spat pronase A fan hitpiournas.pos.orgplesonetarticle7d= 10.137 sournalpore.0100477 ransnaraorr ‘Transcriptomie Analysis of Prunus domestica Undergoing Hypersenstive Response fo Plum Pex Virus Infection 2, Statene 1 2002) Overexpression of plyponal olden trnaginic tomas lat eu inerhanced ctrl oar estnce lan 215: iow Atle » PUBMedNCB! = Google Scholar 3, TongZ, Gao 2, Wang F, Zhu J, Zhang Z (2005) Selection of retale etereace genes fr gene expression tis in geach using fede PCR, BMC Wat Ba 07, hitpiournas.pos.orgplesonetarticle7d= 10.137 sournalpore.0100477 tans.