Resumo

MARTINEZ-MIRAFUENTES, Arturo et al. Predictive capability of DAS-ELISA to barley stripe mosaic virus in wheat. Rev. Mex. Cienc. Agríc [online]. 2016, vol.7, n.6, pp.1321-1333. ISSN 2007-0934.

Detection of pathogens associated with seed requires to count with representative and homogeneous samples to obtain reliable results. This research objective was to evaluate whether: 1) the level of infection (5, 10 and 15%); 2) the increase in the number of single sample (10, 15, 20 and 25) used to form composite samples; and 3) increasing and reducing the usual sample size from 10% (established at CIMMYT for diagnosis of pests and diseases) to 15 and 5%, affect the predictive capability of DAS-ELISA test to barley stripe mosaic virus (BSMV). From seed lots of bread wheat (Triticum aestivum L.), one naturally infected and another healthy, 210 single samples (MS), 20 g each, divided into equal groups of 70 MS samples with 5, 10 and 15% level of infection (NI) to form sample sizes (TM) of 15, 10 and 5%, and composite samples (MC) with 10, 15, 20 and 25 MS samples by NI and TM were formed. The results indicated that reducing the level of NI in MS samples from 15% to 10 and 5% decreased significantly (α= 0.05) in 21 and 31.5% respectively, the sensitivity of the test to BSMV virus. Similarly, reducing the number of MS samples to form MC samples significantly affected (α= 0.05) the sensitivity of the test to the virus and increased the likelihood of false negative record. Although the increase and reduction of TM from 10% to 15 and 5% significantly affected (α= 0.05) the sensitivity of the test to the virus, not finding a linear relationship between TM from the MC sample with the absorbance value. This study demonstrates the importance of the level of infection from the BSMV virus in the seed, of TM and MS number in MS in MC samples to obtain reliable results by DAS-ELISA.

Palavras-chave : BSMV; sampling; seed.