Development of Machine Learning Methods for Accurate Prediction of Plant Disease Resistance

The traditional method of screening plants for disease resistance phenotype is both time-consuming and costly.Genomic selection offers a potential solution to improve efficiency,but accurately predicting plant disease resistance remains a challenge.In this study,we evaluated eight different machine learning(ML)methods,including random forest classification(RFC),support vector classifier(SVC),light gradient boosting machine(lightGBM),random forest classification plus kinship(RFC_K),support vector classification plus kinship(SVC_K),light gradient boosting machine plus kinship(lightGBM_K),deep neural network genomic prediction(DNNGP),and densely connected convolutional networks(DenseNet),for predicting plant disease resistance.Our results demonstrate that the three plus kinship(K)methods developed in this study achieved high prediction accuracy.Specifically,these methods achieved accuracies of up to 95%for rice blast(RB),85%for rice black-streaked dwarf virus(RBSDV),and 85%for rice sheath blight(RSB)when trained and applied to the rice diversity panel I(RDPI).Furthermore,the plus K models performed well in predicting wheat blast(WB)and wheat stripe rust(WSR)diseases,with mean accuracies of up to 90%and 93%,respectively.To assess the generalizability of our models,we applied the trained plus K methods to predict RB disease resistance in an independent population,rice diversity panel II(RDPII).Concurrently,we evaluated the RB resistance of RDPII cultivars using spray inoculation.Comparing the predictions with the spray inoculation results,we found that the accuracy of the plus K methods reached 91%.These findings highlight the effectiveness of the plus K methods(RFC_K,SVC_K,and lightGBM_K)in accurately predicting plant disease resistance for RB,RBSDV,RSB,WB,and WSR.The methods developed in this study not only provide valuable strategies for predicting disease resistance,but also pave the way for using machine learning to streamline genome-based crop breeding.

Identification of stably expressed QTL for resistance to black shank disease in tobacco(Nicotiana tabacum L.) line Beinhart 1000-1

Cigar line Beinhart 1000-1 has effective durable resistance to black shank(BS) and is considered one of the most resistant sources in tobacco(Nicotiana tabacum L.). To investigate the inheritance and identification of stable quantitative trait loci(QTL) for BS response, F2,BC1 F2 individuals and BC1 F2:3 lines were produced from a cross between Beinhart 1000-1 and Xiaohuangjin 1025. Two major quantitative trait loci(M-QTL) named qBS7 and qBS17 were repeatedly detected under different conditions. QTL qBS7 was mapped to the region between PT30174 and PT60621 and explained 17.40%–25.60% of the phenotypic variance under different conditions. The other QTL qBS17 in interval PT61564–PT61538 of linkage group 17 was detected in a BC1 F2 population in the field and in BC1 F2:3 in both the field and at the seedling stage, explaining 6.90% to 11.60% of the phenotypic variance. The results improve our understanding of the inheritance of resistance to BS and provide information that can be used in marker-assisted breeding.

Genome-wide identification and expression analysis of GDSL esterase/lipase genes in tomato

The GDSL esterase/lipase family contains many functional genes that perform important biological functions in growth and development, morphogenesis, seed oil synthesis, and defense responses in plants. The expression of GDSL esterase/lipase genes can respond to biotic and abiotic stresses. Although GDSL esterase/lipase family genes have been identified and studied in other plants, they have not been identified and their functions remain unclear in tomato. This study is the first to identify 80 GDSL esterase/lipase family genes in tomato, which were named SlGELP1–80. These genes were mapped to their positions on the chromosomes and their physical and chemical properties, gene structure, phylogenetic relationships, collinear relationships, and cis-acting elements were analyzed. The spatiotemporal expression characteristics of the Sl GELP genes in tomato were diverse. In addition, RNA-seq analysis indicated that the expression patterns of the SlGELP genes in tomato differed before and after inoculation with Stemphylium lycopersici. qRT-PCR was used to analyze the expression of five Sl GELP genes after treatments with S. lycopersici, salicylic acid and jasmonic acid. Finally, this study was the first to identify and analyze GDSL esterase/lipase family genes in tomato via bioinformatics approaches, and these findings provide new insights for improving the study of plant disease resistance.

Uncovering the dynamic evolution of nucleotidebinding site-leucine-rich repeat(NBS-LRR) genes in Brassicaceae

Plant genomes harbor dozens to hundreds of nucleotide-binding site-leucine-rich repeat （NBS-LRR） genes; however, the long-term evolutionary history of these resistance genes has not been fully understood, This study focuses on five Brassicaceae genomes and the Carica papaya genome to explore changes in NBS-LRR genes that have taken place in this Rosid II lineage during the past 72 million years. Various numbers of NBS-LRR genes were identified from Arabidopsis lyrata （198）, A. thaliana （165）, Brassica rapa （204）, Capsella rubella （127）, Thellungiella salsuginea （88）, and C. papaya （51）. In each genome, the identified NBS-LRR genes were found to be unevenly distributed among chromosomes and most of them were clustered together. Phylogenetic analysis revealed that, before and after Brassicaceae speciation events, both toll/interleukin-1 receptor-NBS-LRR （TNL） genes and non-toll/interleukin-1 receptor-NBS-LRR （nTNL） genes exhibited a pattern of first expansion and then contraction, suggesting that both subclasses of NBS-LRR genes were responding to pathogen pressures synchronically. Further, by examining the gain/loss of TNL and nTNL genes at different evolutionary nodes, this study revealed that both events often occurred more drastically in TNL genes. Finally, the phylogeny of nTNL genes suggested that this NBS-LRR subclass is composed of two separate ancient gene types： RPW8-NBS-LRR and Coiled-coiI-N BS-LRR.

An angiosperm NLR Atlas reveals that NLR gene reduction is associated with ecological specialization and signal transduction component deletion

Nucleotide-binding leucine-rich-repeat(NLR)genes comprise the largest family of plant disease-resis-tance genes.Angiosperm NLR genes are phylogenetically divided into the TNL,CNL,and RNL subclasses.NLR copy numbers and subclass composition vary tremendously across angiosperm genomes.However,the evolutionary associations between genomic NLR content and ecological adaptation,or between NLR content and signal transduction components,are poorly characterized because of limited genome avail-ability.In this study,we established an angiosperm NLR atlas(ANNA,https://biobigdata.nju.edu.cn/ANNA/)that includes NLR genes from over 300 angiosperm genomes.Using ANNA,we revealed that NLR copy numbers differ up to 66-fold among closely related species owing to rapid gene loss and gain.Interestingly,NLR contraction was associated with adaptations to aquatic,parasitic,and carnivorous life-styles.The convergent NLR reduction in aquatic plants resembles the lack of NLR expansion during the long-term evolution of green algae before the colonization of land.A co-evolutionary pattern between NLR subclasses and plant immune pathway components was also identified,suggesting that immune pathway deficiencies may drive TNL loss.Finally,we identified a conserved TNL lineage that may function independently of the EDS1-SAG101-NRG1 module.Collectively,these findings provide new insights into the evolution of NLR genes in the context of ecological adaptation and genome content variation.