Numbers of RFLP probes necessary to show associations between lines

--O.S. Smith, J.S.C. Smith, S.L. Bowen, R.A. Tenborg

Fifty-five pairwise distances ranging from 0.109 to 0.660 were computed from RFLP profiles scored for 125 probes that were well dispersed through the genetic map of maize. In cases where a probe was hybridized against more than one restriction enzyme digest of genomic DNA, only the probe/restriction enzyme digest combination that showed the highest level of polymorphism was utilized for these computations. Genetic distances between lines were calculated (Nei and Li, Proc. Natl. Acad. Sci. USA 76:5256-5273, 1979) for multiple sets of probes in increments of 5 probes. Confidence intervals (CIs) (4x standard error at the 95% level of probability) were computed for pairwise distances between lines. CIs (in parentheses) for the various numbers of probes (not in parentheses) used to generate distances were as follows: 10(0.51), 20(0.37), 30(0.30), 40(0.25), 50(0.23), 60(0.21), 70(0.19), 80(0.18), 90(0.17), 100(0.16), 120(0.15), 140(0.14), 160(0.13), 180(0.12), 200(0.115). Observations of the relative narrowing of CI around the distance value, in terms of the numbers of probes used to generate that distance, allows a more objective decision to be made concerning the numbers of probes to be used for the derivation of such data. When it is important to provide distance data in order to show associations among lines that reflect genetic constitution then the use of 100 probes can provide a reasonable degree of accuracy and is a technically feasible proposition given the current state of RFLP technology.

Return to the MNL 65 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage