Association between relationship measures based on AFLP markers, pedigree data and morphological traits in barley

Thirty one barley lines were used to investigate the agreement between three relationship measures: genetic similarities based on 681 AFLP-markers, coefficients of co-ancestry based on pedigree data, and generalised distance based on 25 morpological characters (morphological distance). Bootstrap analysis was used to estimate the accuracy of the correlation estimates. AFLP-based genetic similarities showed a poor-to-moderate correlation with the coefficients of co-ancestry within the core set of 25 European two-row spring barleys. Morphological distance was not significantly correlated with either genetic similarity or the coefficient of co-ancestry. The precision of all correlation-coefficient estimates, however, was low. The inclusion of two European winter barleys, two American two-row spring barleys, and two American six-row spring barleys in the AFLP-analysis resulted in a much stronger correlation between genetic similarity and the coefficient of co-ancestry. This suggests good opportunities for the use of AFLP-markers to assess genetic diversity by distinguishing between the major ecotypes of barley. Additionally, each of the eight primer combinations used in the AFLP-analysis was able to identify all 31 lines uniquely, showing the usefulness of AFLPs for cultivar identification. Because of the inaccuracy of the investigated relationship measures, resulting in low values of the correlation-coefficient estimates, prediction of the breeding behaviour of parent combinations may be improved by the use of a combination of relationship measures, thus decreasing the effect of their individual independent errors.