Figure 2 - uploaded by Robooni Tumuhimbise
Content may be subject to copyright.
Similar publications
A better understanding of population structure and genetic diversity among cassava germplasm for African cassava mosaic disease and fresh root yield traits is useful for cassava improvement programme. Phenotype-based selection for these traits is cumbersome due to phenotypic plasticity and difficulty in screening of phenotypic-induced variations. T...
Citations
... Bananas are grown by more than 75% of the smallholders in East Africa [ 13] due to their special benefits, which include yields that are adequate despite unpredictable rainfall and their longstanding nature combined with their perennial fruiting nature [ 16]. In terms of consumption per person and total production, as well as the area occupied in Uganda, bananas come in first [ 15]. These qualities set bananas apart as a desirable crop for domestic food production, nourishment, and steady revenue [ 1]. ...
Banana plants are susceptible to the dangerous fungal disease known as Black Sigatoka, which has a negative impact on global economies. Early detection and timely intervention are crucial for preventing the spread of the disease. In recent years, machine learning (ML) has shown great potential for detecting and diagnosing plant diseases, including Black Sigatoka. However, the lack of transparency and interpretability of ML models raises concerns about their use. In this paper, we propose explainable AI approaches for Black Sigatoka detection using Local Interpretable Model-Agnostic Explanations (LIME) and Integrated Gradients. Our methodology involves the utilization of Mobilenet V2 and AlexNet models which are trained on an extensive dataset of banana leaf images and generating explanations to provide a better understanding of the CNN’s decision-making process. We demonstrate the effectiveness of our approach through extensive experiments and show that it outperforms existing state-of-the-art methods for Black Sigatoka detection. Our approach not only provides accurate and interpretable results but also promotes responsible AI practices for plant disease diagnosis.
... This is the final version of the article that is published ahead of the print and online issue Cassava provides food security and income to numerous smallholder farmers in developing countries ( Ceballos et al. 2004;El-Sharkawy 2004;Tumuhimbise 2013;Sewando 2014). Cassava is produced mostly by smallholder farmers on marginal and submarginal lands in Africa. ...
Genetic diversity is essential for crop improvement and knowledge of the genetic variability within a breeding population facilitates its future exploitation. This study examined the phenotypic and genetic variability for specific cassava key traits in a collection of 30 genotypes at five environments in Rwanda. Seventy-three percent of the variation in dry matter content (DMC) and 98% of the variation in total carotene (TC) content was explained by genotypes. The broad-sense heritability (Hbs) of TC was high (99.2%) and the predicted genetic advance (GA; %) from selection was 159.7%. The phenotypic variance coefficients for all traits were higher than the genotypic variance coefficients. The highest Hbs (%) and GA for carotenoids was an indication that conventional breeding could improve carotenoid content in cassava using simple recurrent selection. The postharvest physiological deterioration (PPD) showed significant (p < 0.05) negative correlation and interrelation with TC and DMC, indicating that the high-TC and low-DMC cultivars could have delayed PPD. The higher-TC genotypes could be used to improve TC in the cassava population and consequently could contribute to the development of delayed-PPD cultivars. The results confirm that there is useful variation for key traits within the collection that could be indicative of a broad range of useful variability in natural populations, landraces and cultivars that could be exploited for breeding.
... The potential yield of cassava is estimated at 90 t ha −1 of fresh roots under well managed conditions (El-Sharkawy 2004). Cassava plays a key role as a food security and income-generating crop for many smallholder farmers in developing countries (Ceballos et al. 2004; El-Sharkawy 2004; Tumuhimbise 2013). In East Africa, cassava is eaten after boiling and processing to flour to make porridge, local brew, ugali and bread, and sweet varieties lacking cyanogenic glycosides can be eaten raw (Kamau 2006; Mkumbira et al. 2003; Were 2011). ...
... In addition, relying on varietal introduction , participatory varietal selection is limited in Rwanda. On the other hand, the current research conducted in East Africa shows that the limited involvement of end-users in the formal breeding process affected the level of adoption of new cultivars negatively (Kamau et al. 2011; Tumuhimbise 2013; Were et al. 2014). Many breeding programmes in developing countries fail due to the lack of inclusion of participatory approaches which negatively affects the level of adoption of newly developed cultivars (Kamau et al. 2011; Were et al. 2012). ...
Physiological postharvest deterioration (PPD) and late bulking are among the traits that make cassava an unattractive crop in many environments. This study aimed at assessing the main constraints of cassava production, the effects of late bulking, the losses due to PPD and the factors affecting adoption of new cultivars in Rwanda. A participatory rural appraisal (PRA) and a baseline survey were conducted in March-May 2014 in three agro-ecological zones in the country using a multistage sampling method. Cassava was grown on 0.29 ha out of 0.69 ha total average land possession per household. The majority of cassava farmers (59.1 %) practised intercropping as their land holding is small. Average yield was 21.8 t ha−1. A number of constraints was identified, particularly the lack of clean cuttings, viral diseases, late bulking cultivars, drought, limited information and knowledge, weathered soils, insufficient fertilizers, land shortage, lack of markets and effective storage techniques. Loss due to PPD was estimated at 11.9 % of total production per year. Piecemeal harvest and underground storage of roots were the main practices used to delay PPD. Change in colour and taste, rotting, difficulty in removing skin and increase of fibres in the flesh were the farmers’ methods for assessing PPD. Time to harvest varied from district to district and was attributed to genetic x environment interactions. The use of late bulking varieties and the lack of yield production of other crops resulted in reduced food availability and potential food crises. Farmer preferences, information and extension services, performance, quality, market acceptability and cutting production influenced the adoption of new cassava cultivars. Thus, breeding objectives targeting the end user preferences could enhance the adoption of new cultivars.
Cassava brown streak disease (CBSD) threatens cassava production in sub-Saharan Africa despite the availability of resistant varieties. Extreme environmental factors weaken plant defenses, reducing CBSD resistance. This study examined CBSD inheritance in cassava populations, assessed genetic variability, and identified superior sources of resistance using F1, S1, and half-sib offspring populations derived from resistant sources. The offspring underwent field evaluation at two distinct sites from 2019 to 2021, and the symptom-free genotypes were analyzed using reverse transcription-quantitative polymerase chain reaction (RT-qPCR). Resistance to CBSD was categorized as most resistant, resistant, most tolerant, or tolerant based on symptoms and virus titers. The findings indicated that the resistance to CBSD is highly influenced by genotypes, F1/S1 types, and environmental conditions. An analysis of combining abilities revealed significant general combining abilities (GCAs) for CBSD, cassava mosaic disease (CMD), and traits associated with yield. The heritability estimates for resistance to CBSD varied between 43.4% and 63.2% for foliar symptoms and 14.6% and 57.9% for root necrosis across locations. The inheritance pattern involved a combination of additive and recessive genes with selfed (S1) populations displaying stronger and more effective resistance to the disease. The cassava brown streak virus (CBSV) was highly prevalent, and the Ugandan cassava brown streak virus (UCBSV) was not prevalent. Four genotypes were highly resistant to CBSD and could be key sources of resistance to this disease.
Cassava (Manihot esculenta Crantz) brown streak disease (CBSD) is currently one of the major constraints to sustained cassava production in Malawi. Its economic impact is mainly manifested in the storage roots where it causes pitting, necrosis and constriction. CBSD can effectively be managed by using resistant varieties as well as early harvesting especially if the varieties are early bulking (high storage root yield). However, development of resistant and high storage root bulking varieties requires an understanding of gene action controlling inheritance of the two traits. Currently, there is very little information in Malawi regarding the inheritance pattern and relative importance of general (GCA) and specific combining ability (SCA) of these two traits. Therefore, a study was conducted to determine mode of gene action as well as importance of GCA and SCA in the genetic control of CBSD resistance and storage root yield. Information generated is essential in the selection of parents and breeding strategies for an effective breeding programme. Thirty-six crosses were generated using a 6 × 6 North Carolina Design II and later evaluated for CBSD resistance and storage root bulking at two locations using a triple square lattice design. Data for the progenies were analysed using REML in Genstat (17th ed.) while family (crosses) data were analysed using a PROC Mixed procedure performed in SAS® 9.3 Software. The GCAf × location, GCAm × location and SCA × location interaction effects were highly significant (P < 0.01) for CBSDS and CBSDI. Pooled analysis across two locations showed highly significant (P < 0.01) variance among GCA due to females (GCAf) and GCA due to males (GCAm) for FSRY. These GCAf, GCAm and female × male interaction (SCA) effects accounted for 19.8, 33.4 and 46.7% of the sum of squares for CBSDS, respectively. The results suggested that additive gene effects predominantly controlled CBSD resistance and high storage root bulking. Four parental genotypes (Silira, Mulola, Phoso and Mkondezi) were identified as the best general combiners for the CBSD, high storage root bulking and other traits. Thirteen progenies exhibiting CBSD resistance and high storage root bulking were identified and selected for advancement.
Background:
The early generation selection of cassava quantitative and qualitative traits saves breeding resources as it can shorten breeding schemes. The inheritance analysis provides important breeding information for developing new improved varieties. This study aimed at developing an F1 segregating cassava population and determining mode of gene action of pulp colour and selected traits at early generation selection (F1 seedling and clones).
Results:
The fifteen families exhibited significant (p<0.05) phenotypic variation between offspring. The general combining ability (GCA) was significant for all traits except cassava brown streak disease on leaves, while specific combining ability (SCA) was significant for all evaluated traits. The Garukansubire and Gitamisi were the best general combiners for improving fresh storage root yield while G1 and G2 were the best general combiners for improved carotenoid (yellow/ orange pulp colour) and delayed physiological postharvest deterioration. The pulp colour had the highest GCA/ SCA ratio and % sum of square (SS) due to GCA.
Conclusion:
The 15 F1 families exhibited essential genetic diversity for cassava improvement. The expression of most cassava traits was controlled by both additive and non-additive gene action. The study elucidated the role of dominance effects over the additive effects for the evaluated traits. However, the pulp colour was predominantly controlled by additive gene action. This implies the possibility of improving cassava through conventional breeding using recurrent selection for most traits.