Gloria Valentine Nakato

Banana breeding programs prioritize improving yield with resistance to biotic stresses while maintaining fruit quality. This study focused on identifying genotypes with resistance to five major biotic constraints including Cosmopolites sordidus , Radopholus similis , banana Xanthomonas wilt, Fusarium oxysporum f. sp. cubense – Race 1 and Pseudocercospora fijiensis . A diverse set of diploid and tetraploid genotypes was phenotyped under controlled screenhouse and field conditions to identify potential sources of. A partially replicated (P-rep) design, was used to accommodate unequal plant availability arising from asynchronous multiplication. P-rep design enabled robust phenotypic comparisons and identification of genotypes expressing resistance to multiple biotic stresses. Several diploid genotypes exhibited resistance to multiple constraints, highlighting their potential as parental lines for resistance introgression and broadening the genetic base for banana improvement. Leveraging on the efficiency and precision of the P-rep experimental design, these findings accelerate the phenotyping of genotypes.

Bacterial diseases of banana are a growing global threat, causing yield losses and increased management costs. Major diseases include Moko, banana blood disease (BBD), and banana Xanthomonas wilt (BXW), caused by Ralstonia solanacearum , Ralstonia syzygii subsp. celebesensis , and Xanthomonas vasicola pv. musacearum , respectively. Effective surveillance requires point-of-care diagnostics such as loop-mediated isothermal amplification (LAMP) for on-site use. We aimed to develop three LAMP assays to specifically detect the bacteria responsible for Moko, BBD and BXW, directly from banana tissues, using a simplified DNA extraction protocol. The BBD – and BXW-LAMP assays demonstrated 100% specificity, yielding negative results for a broad range of non-target bacteria, including closely related species as well as pathogenic and endophytic strains associated with banana, and positive results for all the tested target strains. For Moko disease, a duplex-LAMP assay was developed to detect all strains from the four globally most relevant sequevars: IIB-3, IIB-4, IIA-6, and IIA-24. The duplex-LAMP successfully detected all target strains, except one that was shown to be non-pathogenic to Cavendish bananas. All non-target strains tested negative, with the exception of a delayed signal for one strain belonging to Ralstonia thomasi , not associated with banana environment (hospital strain). These results were supported by an extensive in silico analysis conducted on 9,668 Burkholderiaceae and 7,483 Xanthomonadaceae genomes. Detection limits ranged from 0.1 pg/µl to 1 pg/µl DNA, and from 10 ⁴ to 10 ⁵ CFU/ml on banana tissues spiked with calibrated bacterial suspensions, depending on the assay. The LAMP assays prove highly effective for detecting target pathogens in both artificially inoculated banana plants and field samples, offering a promising tool for improving disease management strategies.

Accurately assessing weevil damage is critical when evaluating banana germplasm. This enables identification of genotypes resistant to the banana weevil, Cosmopolites sordidus , for use as elite parents or for advancement in banana breeding programs. Although visual observation remains the most common phenotyping approach, it is limited by individual bias. This study investigated the potential of image analyses as precise and objective alternatives for assessing weevil damage on the banana corm. ImageJ and four machine learning models, to include YOLO_v11 medium, U-Net 512, SegFormer, and MiT-b0, were explored. 260, 65, and 72 images were used to train, test, and validate the machine learning models, respectively. Phenotyping trials were set up as partially replicated (P-rep) designs with 18 test genotypes and four control genotypes. All plants were produced through tissue culture, raised in pots before infestation with the weevils. At termination, the percentage score of each of the 370 corm samples was evaluated both visually and by image analysis to compare scores across all methods. A significant genotype effect was detected, indicating differences among genotypes in resistance to banana weevils. Furthermore, a significant genotype-by-scoring-method interaction showed that genotype rankings in terms of resistance to banana weevils varied across methods. This emphasized that the choice of scoring approach can affect the magnitude of damage quantified. Visual observation agreed more closely with image analyses for smaller scores and less for larger scores. Results from genotype performance evaluation showed that machine learning methods, except for the YOLO_v11, have a strong level of agreement and can be used interchangeably, giving consistent, reliable, and repeatable measurements. We recommend and have adopted machine learning when scoring weevil damage in the banana corm to avoid individual bias and subjectivity arising from visual observation.

Bacterial diseases of banana are becoming increasingly significant worldwide, resulting in reduced yields and higher disease management costs. The most important bacterial diseases of banana include Moko and banana blood disease (BBD), caused by Ralstonia solanacearum and Ralstonia syzygii subsp. celebesensis , respectively, and banana Xanthomonas wilt (BXW) caused by Xanthomonas vasicola pv. musacearum . Effective surveillance and disease management require point-of-care diagnostics, such as loop-mediated isothermal amplification (LAMP), for on-site operation. In this study, three LAMP assays were developed to specifically detect the bacteria responsible for Moko, BBD and BXW, directly from banana tissues, using a simplified DNA extraction protocol. The BBD - and BXW-LAMP assays demonstrated 100% specificity, yielding negative results for a broad range of non-target bacteria, including closely related species as well as pathogenic and endophytic strains associated with banana, and positive results for all the tested target strains. For Moko disease, a duplex-LAMP assay was developed to detect all strains from the four globally most relevant sequevars: IIB-3, IIB-4, IIA-6, and IIA-24. The duplex-LAMP successfully detected all target strains, except one that was shown to be non-pathogenic to Cavendish bananas. All non-target strains tested negative, with the exception of a delayed signal for one strain belonging to Ralstonia thomasi , not associated with banana environment (hospital strain). These results were supported by an extensive in silico analysis conducted on 9,668 Burkholderiaceae and 7,483 Xanthomonadaceae genomes. Detection limits ranged from 0.1 pg/µl to 1 pg/µl DNA, and from 10⁴ to 10⁵ CFU/ml on banana tissues spiked with calibrated bacterial suspensions, depending on the assay. The LAMP assays prove highly effective for detecting target pathogens in both artificially inoculated banana plants and field samples, offering a promising tool for improving disease management strategies.

Radopholus similis severely damages banana roots causing significant yield losses. Field screening for resistance is labor intensive and inconsistent due to environmental variation and mixed nematode populations. The screenhouse offers a controlled environment but is limited by the time needed for root development and variation in plant growth. We developed and validated a high-throughput in vitro method for phenotyping banana resistance to R. similis using sand-Murashige and Skoog (MS) media. Tissue culture plantlets grown in sterilized sand-MS were inoculated with 50 female R. similis after root development and nematodes extracted eight weeks after inoculation to calculate the reproduction factor (RF). Although RF values were higher for in vitro than in the screenhouse, accession responses showed similar trends under both conditions. The in vitro method was rapid, cost-effective with higher throughput, accelerating phenotyping and enabling rapid assessment of banana accessions for breeding programs. Some accessions responded differently to the two methods indicating that additional methods, such as root necrosis scores are important to confirm resistance. This study is the first in vitro-based demonstration of phenotyping for nematode resistance using modified sand-MS media with improved root development and pathogen interactions.

CONTEXT: Tropical cropping systems must adapt to the current and future geographic distribution of pathogen and pest communities. An important research gap is how climate change may shift the distribution of pathogens and pests in tropical lowlands and highlands. OBJECTIVE: We evaluated the current geographic risk of 27 pathogens and pests in the production of four food security crops (banana, cassava, potato, and sweetpotato) in the Great Lakes region of Africa, and the potential future risk under climate change. Models for each pathogen and pest indicate the potential for changes in geographic distribution, with model fit indicating the potential for decision support systems to facilitate management. METHODS: First, cropland connectivity analysis identified locations likely important in the spread of crop-specific pathogens and pests, such as locations in Rwanda and Burundi. Second, we surveyed the 27 economically important pathogens and pests in Rwanda and Burundi, mapping the distribution of each across climate gradients and quantifying patterns of association. Third, we used machine learning to develop models of each species as a function of environmental variables, including host landscape variables. We also evaluated the increase in temperature across altitudes under future climate change scenarios in this region. RESULTS AND CONCLUSIONS: Among the ten machine-learning algorithms evaluated, random forests and support vector machines generally performed best for predicting severity and infestation. Host landscape variables were useful predictors for some species. Based on climate matching, 44% of the pathogens and pests could become more common with warmer temperatures at higher altitudes, while 17% may become less common. SIGNIFICANCE: This study indicates adaptation priorities for crop health in a region with multiple challenges to agricultural sustainability. The models developed here also indicate which species may have more potential and relevance for future development of pathogen and pest forecasts.

Accurately assessing weevil damage is critical when evaluating banana germplasm to identify genotypes resistant to the banana weevil (Cosmopolites sordidus), for use as elite parents in the banana breeding pipeline or evaluating breeding products. Visual observation remains the most common phenotyping approach but limited by individual bias. This study investigated the potential of image analyses as precise and objective alternatives for assessing weevil damage on the banana corm. Phenotyping trials were set up as partially replicated (P-rep) designs with 22 tissue culture-generated genotypes raised in pots and infested with banana weevils. At termination, the percentage score of weevil damage to the corms was evaluated by visual observation and image analysis using ImageJ and machine learning. In total, 370 high-quality images were assessed for weevil damage using ImageJ and machine learning. On average, damage scores from visual observation were 5.52% and 3.88% higher than ImageJ and machine learning respectively. There was a proportional trend with visual observation agreeing closely to image analyses for smaller scores and but less for larger scores. The results show that both ImageJ and machine learning exhibited a strong level of agreement and are interchangeable with consistent, reliable, and repeatable measurements. In conclusion, to avoid individual bias and subjectivity arising from visual observation, we recommend the use of either ImageJ or machine learning when scoring weevil damage in the banana corm.

Banana Xanthomonas wilt, caused by Xanthomonas vasicola pv. musacearum (Xvm), is a devastating disease that results in total yield loss of affected plants. Resistance to the disease is limited in Musa acuminata, but it has been identified so far in the zebrina subspecies. This study identified markers associated with tolerance to Xvm in Monyet, a tetraploid banana from the zebrina subspecies which was identified to be partially resistant to the bacterium. We used a triploid progeny of 135 F₁ hybrids resulting from a cross between Monyet (Xvm partially resistant) and Kokopo (diploid and Xvm susceptible). The F₁ hybrids were screened in pots for resistance to Xvm. The population was genotyped using the genotyping-by-sequencing platform of Diversity Array Technology (DArTSeq). The adjusted means of the phenotypic data were combined with the allele frequencies of the genotypic data in continuous mapping. We identified 25 SNPs associated with resistance to Xvm, and these were grouped into five quantitative traits loci (QTL) on chromosomes 2, 3, 6, and 7. For each marker, we identified the favorable allele and the additive effect of replacing the reference allele with the alternative allele. The comparison between weevil borer (Cosmopolites sordidus (Germar)) and Xvm QTL revealed one QTL shared between the two biotic stresses at the distal end of chromosome 6 but with a repulsion linkage. This linkage should be broken down by generating more recombinants in the region. We also identified 18 putative alleles in the vicinity of the SNPs associated with resistance to Xvm. Among the 18 putative genes, two particularly putative genes, namely, Ma06_g13550 and Ma06_g36840, are most likely linked to disease resistance. This study is a basis for marker-assisted selection to improve banana resistance to banana Xanthomonas wilt, especially in East and Central Africa where the disease is still devastating the crop.

For decades, Xanthomonas vasicola pv. musacearum (Xvm) has been an economically important bacterial pathogen on enset in Ethiopia. Since 2001, Xvm has also been responsible for significant losses to banana crops in several East and Central African countries, with devastating consequences for smallholder farmers. Understanding the genetic diversity within Xvm populations is essential for the smart design of transnationally reasoned, durable, and effective management practices. Previous studies have revealed limited genetic diversity in Xvm, with East African isolates from banana each falling into one of two closely related clades previously designated as sublineages SL 1 and SL 2, the former of which had also been detected on banana and enset in Ethiopia. Given the presumed origin of Xvm in Ethiopia, we hypothesized that both clades might be found in that country, along with additional genotypes not seen in Central and East African bananas. Genotyping of 97 isolates and whole‐genome sequencing of 15 isolates revealed not only the presence of SL 2 in Ethiopia, but additional diversity beyond SL 1 and SL 2 in four new clades. Moreover, SL 2 was detected in the Democratic Republic of Congo, where previously SL 1 was the only clade reported. These results demonstrate a greater range of genetic diversity among Xvm isolates than previously reported, especially in Ethiopia, and further support the hypothesis that the East/Central Africa xanthomonas wilt epidemic has been caused by a restricted set of genotypes drawn from a highly diverse pathogen pool in Ethiopia.