Back to CGN

APIC format for potato evaluation data 


Short summary of:

Hoekstra, R., J.B. Bamberg & Z. Huamán, 1997. A case study on merging evaluation data from different genebanks: the Inter-genebank Potato Database. Pp. 69-71 in: Central Crop Databases: Tools for Plant Genetic Resources management. Report of a workshop, 13-16 October 1996, Budapest, Hungary (E.Lipman, M.W.M. Jongen, Th.J.L. van Hintum, T. Gass and L. Maggioni, eds.). IPGRI/CGN.

The comparison of evaluation data from different sources is difficult due to differences in the evaluation methods and environments. The world potato genebanks, members of APIC (the Association of Potato Intergenebank Collaborators), agreed upon standard descriptors and descriptor states, describing the reaction of the plants to stresses (diseases and pests) and their quality traits. Characterization data such as plant height were not included.

Since most evaluation data on wild potatoes are the results of screening populations and the data obtained from different sources, plant breeders can use the information only as a general guide for selecting germplasm. The quality of the data must be validated by the users by selecting individual plants.

Following standard codes were adopted:

Pests / diseases

Quality traits

code meaning code meaning
VR Very Resistant VH Very High
R Resistant H High
M interMediate M interMediate
S Susceptible L Low
VS Very Susceptible VL Very Low
T Tolerant
H Hypersensitive numeric values
I Immune

Sometimes additional information to the evaluation results is available. Therefore, two extra digits were included next to the two characters representing the reaction (VR, R, M, S, VS):

Examples

Another convention adopted by APIC was that when very few good sources for resistance are available or the valuable trait is rare, for example resistance in a minor portion of the population, then the reaction will be coded for example as R1 instead of S9. The genebank curator in cooperation with the experts who carried out the evaluations make this decision and transform the data into the common format.

Conclusions

It is essential to use descriptor states that can be accommodated to data generated by different researchers, in different environments, and with different evaluation methods. As much as possible there should be access to the descriptions of the screening method and environment where the material was screened. The descriptor states should be more or less of general use and preferably should include those that are more or less self-explanatory (i.e., resistant, susceptible, etc.).


Feedback to: Lana.deBruijn@wur.nl,
Centre for Genetic Resources the Netherlands (CGN), Wageningen, the Netherlands.