DataSheet_1_Alternative Approaches for Assessing Cassava Brown Streak Root Necrosis to Guide Resistance Breeding and Selection.xls

Cassava brown streak disease (CBSD) caused by the rapidly evolving cassava brown streak viruses (CBSVs), causes immense yield losses to the cassava value chain in eastern and southern Africa. Western Africa, another region that heavily depends on cassava is under eminent threat from CBSD. Resistance breeding is the best practical solution. However, complexities associated with CBSD resistance screening i.e., variable root sampling units, limit systematic attainment of genetic progress. Accordingly, we compared efficiency of five CBSD root necrosis assessment methods to guide selection: cassava brown streak disease root incidence (CBSDRi), cassava brown streak disease root severity (CBSDRs), cassava brown streak disease root severity computed as harmonic mean (CBSD-Harmonic), proportion-based root necrosis index (CBSD-proportion), and standardized root necrosis index (CBSD-standardized). The indexes (CBSD-proportion and CBSD-standardized) correct for variable sample size. We analyzed CBSD root necrosis data of 256 clones evaluated across 12 environments. Higher and variable standard errors were associated with root severity score 1 (no CBSD root necrosis). Lowest and highest plot-based heritability were respectively registered for CBSD-standardized (0.22) and CBSD-proportion (0.71). CBSDRs was only positively correlated with CBSDRi (r = 0.92) and CBSD-Harmonic (r = 0.97). Using best linear unbiased predictions (BLUPs), we ranked the top 15 CBSD resistant clones; only one clone (UG130014) featured in all the five assessment methods; two clones (UG130006 and UG120156) featured in four (CBSD-Harmonic, CBSDRi, CBSDRs, and CBSD-standardized); and five clones (UG120180, UG120063, UG130002, UG130033, and UG120183) featured in three methods (CBSD-Harmonic, CBSDRi, and CBSDRs). Influence of sample size was also quantified by sub-setting and analyzing CBSDRs data to have plots with at least 40 or 30 roots. Data stabilization was evident in plots with 30 roots. The significant influence of root sample sizes on overall ranking of clones, justifies the use of CBSD root necrosis indexes in early selection stages i.e., seedling and/or clonal trials, that are often characterized by high variations in roots assessed per plot. It is expected that this information will provide a foundation for harmonizing and/or optimizing on-going and future CBSD resistance breeding efforts.

由快速演化的木薯褐条病毒（cassava brown streak viruses, CBSVs）引发的木薯褐条病（Cassava brown streak disease, CBSD），给东非与南非的木薯产业链造成了巨额产量损失。作为另一高度依赖木薯的地区，西非同样面临CBSD的严峻威胁。抗病育种是当前最切实可行的防控解决方案。然而，木薯褐条病抗病筛选面临的核心难题之一为根部取样单元不统一，这制约了遗传改良进程的系统性推进。为此，本研究对比了5种CBSD根部坏死病评估方法的效率，以指导最优筛选方案的选择：木薯褐条病根部发病率（cassava brown streak disease root incidence, CBSDRi）、木薯褐条病根部病情严重度（cassava brown streak disease root severity, CBSDRs）、以调和均值计算的木薯褐条病根部病情严重度（cassava brown streak disease root severity computed as harmonic mean, CBSD-Harmonic）、基于比例的根部坏死指数（proportion-based root necrosis index, CBSD-proportion）以及标准化根部坏死指数（standardized root necrosis index, CBSD-standardized）。其中CBSD-proportion与CBSD-standardized这两个指数可校正取样量的差异。本研究分析了12个试验环境下种植的256个木薯无性系的CBSD根部坏死病数据。结果显示，病情严重度评分1级（无根部坏死症状）对应的标准误更高且变异度更大。基于小区的遗传力以CBSD-standardized最低（0.22），CBSD-proportion最高（0.71）。CBSDRs仅与CBSDRi（r=0.92）及CBSD-Harmonic（r=0.97）呈显著正相关。利用最佳线性无偏预测（best linear unbiased predictions, BLUPs），本研究对15个高抗CBSD的木薯无性系进行了排名：仅UG130014这1个无性系在全部5种评估方法中均入选；UG130006与UG120156这2个无性系在4种方法（CBSD-Harmonic、CBSDRi、CBSDRs及CBSD-standardized）中均入选；另有UG120180、UG120063、UG130002、UG130033及UG120183这5个无性系在3种方法（CBSD-Harmonic、CBSDRi及CBSDRs）中入选。本研究还通过对CBSDRs数据进行子集划分与分析——仅保留单小区至少包含40或30个样本的分组——量化了根部取样量的影响。结果发现，当单小区样本量为30时，数据稳定性显著提升。根部取样量对无性系综合排名存在显著影响，这证明在早期选择阶段（即苗期试验和/或无性系试验）采用CBSD根部坏死指数是合理的——这类试验通常存在单小区评估样本量差异较大的问题。本研究结果可为当前及未来的CBSD抗病育种工作的标准化与优化提供坚实的理论基础。