CIMMYT Publications Repository
Not a member yet
7809 research outputs found
Sort by
Optimizing genomic prediction with transfer learning under a ridge regression framework
No full textGenomic selection (GS) is a predictive plant and animal methodology that allows the selection of plants and animals based on predictions without the need to measure the phenotype. However, its practical application requires challenging prediction accuracy due to the noise observations collected in experiments in these areas. Many strategies and approaches have been proposed to improve the prediction accuracy of this methodology. This paper explores the use of transfer learning in the context of GS. Transfer learning with (1) ridge regression (RR) (Transfer RR) and (2) analytic RR (ARR) (Transfer ARR) were applied from cultivars in the proxy environment to predict those cultivars in the goal environments. Also, we compared the performance of models RR and ARR without transfer learning. We used 11 real multi-environment datasets (wheat and rice) and evaluated them in terms of Pearson's correlation (Cor) and normalized root mean square error (NRMSE). Our study shows empirical evidence that the Transfer RR or Transfer ARR approaches significantly enhanced predictive performance. Across the datasets, Transfer RR (or Transfer ARR) method improved Cor by 22.962% and NRMSE by 5.757%, in comparison to models RR and ARR. These results underscore the potential of Transfer RR (or Transfer ARR) when enhancing predictive accuracy in this context
Aflatoxins in Mexican maize systems: From genetic resources to agroecological resilience and co-occurrence with fumonisins
No full textAflatoxins (AFs) and fumonisins (FUMs) are among the most prevalent and toxic mycotoxins affecting maize production globally. In Mexico, their co-occurrence poses a significant public health concern, as maize is not only a dietary staple but also predominantly grown and consumed at the household level. This review examines the multifactorial nature of AFs and FUMs contamination in Mexican maize systems, considering the roles of maize germplasm, agricultural practices, environmental conditions, and soil microbiota. Maize landraces, well-adapted to diverse agroecological zones, exhibit potential resistance to AFs contamination and should be prioritized in breeding programs. Sustainable agricultural practices and biocontrol strategies, including the use of atoxigenic Aspergillus flavus strains, are presented as promising interventions. Environmental factors and soil characteristics further influence fungal proliferation and mycotoxin biosynthesis. Advances in microbiome engineering, biological breeding approaches, and predictive modeling offer novel opportunities for prevention and control. The synergistic toxicity of AFs and FUMs significantly increases health risks, particularly for liver cancer, highlighting the urgency of integrated mitigation strategies. While Mexico has regulatory limits for AFs, the lack of legal thresholds for FUMs remains a critical gap in food safety legislation. This comprehensive review underscores the need for biomarker-based exposure assessments and coordinated national policies, alongside multidisciplinary strategies to reduce mycotoxin exposure and enhance food safety in maize systems
Sustainability effects of the uptake of more grass-based feeding practices: evidence from Sweden
No full textWe examine the effects of the uptake of grass-based feeding practices on the economic, environmental, and social dimensions of farm sustainability. More specifically, we analyse the predictive effects of an increase in grassland or ley area on farm net income, total working hours, and fertiliser expenditure. Our analysis utilises farm-level data from Swedish dairy farms spanning the period 2002–2021. Drawing on a directed acyclic graph (DAG), we assess both the average and distributional effects using panel regression with fixed effects and penalised panel quantile regression methods, respectively. Our results show that an increase in either grassland or ley area is associated with a decrease in both farm net income and fertiliser expenditure, alongside an increase in total working hours on average. Our distributional analysis further indicates that these effects vary across the quantile distribution of the outcome variables. Finally, the results show that an increase in either grassland or ley area leads to a reduction in milk yield, feed cost and the cost of veterinary services. Overall, our findings highlight the trade-offs associated with farmers' transition towards the uptake of more grass-based feeding practices and the policy implications
Impact of irrigation pump ownership on farm productivity in rice-wheat cropping systems of Nepal Terai
No full textGroundwater irrigation is critical for supporting food security, rural livelihoods, and economic development in the Eastern Indo-Gangetic Plains (EIGP) of South Asia. However, groundwater resources in the EIGP remain underutilized due to farmers' reliance on expensive diesel pumps for accessing water. This study uses primary household survey data from the Terai region of Nepal to analyse the drivers of variability in irrigation access costs across farms, and how these cost variabilities influence agricultural outcomes. We employ an endogenous switching regression model to assess the impacts of pump ownership on farm productivity and profitability. Our findings show that pump ownership reduces irrigation costs rice and wheat cultivation, the region's two major crops, by 72 % and 76 %, respectively, and increases rice and wheat productivity by 37 % and 20 %, respectively. Our findings provide empirical evidence of the positive impact of technology ownership on agricultural productivity and highlight the opportunities for policy interventions focused on improving performance of existing technologies for enabling long-term sustainable intensification of irrigated agriculture in the EIGP
Assessing herbicide use and hand weeding efficacy in groundnut production intensification
No full textPoor and costly weed management constrains Groundnut (Arachis hypogea L.) production in Uganda. A field|study was therefore conducted at the National Semi-Arid Resources Research Institute (NaSARRI), Serere,Uganda during the long rains of 2020 and 2021 and short rains of 2020 to evaluate the efficacy of hand weeding and different herbicides on weed management, yield, and the economics of their use in groundnut. The experiment for this study comprised 7 treatments constituted by six herbicides; four pre-emergent (Glyphosate, Clethodim, S-Metolachlar, and 2,4-Dichlorophenoxyacetic acid), and two post-emergent (Bentazone and Quizalofop-p-ethyl) and hand weeding. Post-emergence herbicide application and hand weeding were done at 30, 45, and 60 DAS. The treatments were laid out in a randomized complete block design (RCBD) with three replications. Calculated weed indices show the effect of weed control measures on groundnut weeds. Pre-emergence application of glyphosate followed by post-emergence application of Quizalofop-p-ethyl produced superior pod yield (1724.3 kg/ha), the lowest weed density of grass (0.62), and Sedges (0.61), the lowest weed biomass at harvest (122.5g), the highest percentage of weed control efficiency (69.65%), and highest net returns (7,937,746UGX/ha). However, post-emergence sole application of quizalofop-p-ethyl produced the highest B: C ratio (36.49). Therefore, this study has indicated that the pre-emergence application of glyphosate followed by the post-emergence application of quizalofop-p-ethyl is the most profitable weed control measure in groundnut; while the post-emergence sole application of quizalofop-p-ethyl is the most economical. Hand weeding though may be used where labour is cheap and not scarce as opposed to the herbicides.69-8
Fine mapping and candidate gene analysis of a major QTL for grain length on chromosome 5BS in bread wheat
No full textLarge grain is a favorable trait for appearance quality and a large sink potential in wheat breeding. The stable QTL QGl.caas-5BS for grain length was previously identified in a recombinant inbred line population from the cross of Zhongmai 871 (ZM871) and its sister line Zhongmai 895 (ZM895). Here, a BC1F6 residual heterozygous line was selected from the cross of a ZM871/ZM895//ZM871 population, and six heterozygous recombinant plants were identified in the BC1F7 population from self-pollination of the heterozygous line. QGl.caas-5BS was delimited into an interval of approximately 2.2 Mb flanked by markers Kasp_5B33 and Kasp_5B2 (25.3–27.5 Mb) by phenotyping and genotyping the secondary mapping populations derived from these heterozygous recombinant plants. Five genes were predicted as candidates of QGl.caas-5BS based on sequence polymorphism and differential expression analyses. Further mutation analysis showed that TraesCS5B02G026800 is likely the causal gene of QGl.caas-5BS. The gene-specific marker Kasp_5B_Gl for TraesCS5B02G026800 was developed, and a significant genetic effect of QGl.caas-5BS on grain length was identified in a validation population of 166 cultivars using this marker. These findings lay a good foundation for map-based cloning of QGl.caas-5BS and provide a breeding-applicable marker for the improvement of grain length in wheat.2465-247
Impact of wheat production technology packages adoption on smallholder farmers’ food security and income in Horo Guduru Wollega Zone, Ethiopia
No full textAdoption of agricultural technology packages got considerable attention in enhancing smallholders’ food security and farm income in Ethiopia. However, the impact evaluation of such technology packages’ contributions to households’ food security and income are limited. This study is aimed at identifying factors affecting households’ decisions in adopting wheat technology packages and its ex-post adoption impact on their food security and wheat production income. The food consumption score and households’ dietary diversity score were used as proxy measures of households’ food security. The study employed a multinomial endogenous switching regression model to account for selection bias. Analysis results show that household decisions to adopt combinations of wheat technology packages influenced by age, sex, education level of household head, distance from markets, plots and services, phone ownership, access to credit services, agricultural cooperative membership, farm size, livestock, and farm training services. Moreover, the study shows that adoption of full wheat technology packages has a greater positive impact on households’ food security and wheat production income than adopting these packages in a few combinations or in isolation. The findings of the study suggest strengthening the provision of improved agricultural services to support farmers in adopting packages of technologies for a better food security and livelihood outcomes
Impact of seed producer cooperatives membership on technical efficiency: evidence from tef farmers in the central highlands of Ethiopia
No full textFarmers in Ethiopia face challenges associated with low efficiency and productivity, primarily due to limited access to improved seeds and complementary inputs. Despite empirical evidence of the vital role played by agricultural cooperatives in providing these basic inputs, there has been no empirical study examining the impact of membership in seed producer cooperatives (SPCs) on agricultural development. This study aimed to assess the impact of membership in SPC on the technical efficiency (TE) of farmers in the central highlands of Ethiopia. Using data from 425 selected tef farmers, the paper employed propensity score matching to match SPC members with non-members and applied a selectivity-corrected stochastic production frontier model to address unobserved biases. In addition, the meta-frontier approach was used to compare TE scores between the two groups of farmers. Results reveal that SPC members achieved a TE of 72 % of their potential output, whereas non-members achieved only 59 %. SPC members achieved meta technical efficiency (MTE) of 67 %, while non-members obtained MTE of 49 %. This indicates that non-members face tef production challenges due to limited access to improved technologies. In conclusion, SPC membership significantly enhances the TE of members compared to non-members indicating that improving TE could greatly boost productivity in the tef sector. The findings of the study suggest that the government of Ethiopia should enhance the skills of SPC members by providing training and promoting knowledge sharing among farmers to improve seed access. Similarly, private seed producers should be strengthened to meet the growing demand for improved seeds