INTRODUCTION
 
Plant Introductions from centers of crop diversity have been valuable source materials for many traits, including host-plant resistance to biotic and abiotic stresses. This genetic diversity has resulted from evolutionary processes; including mutation, recombination, natural selection, genetic drift, and migration in the many ecogeographic niches. Human intervention has increased the diversity, but seed exchange during trade and migration tends to reduce diversity among ecogeographic niches.
 
When scientists cannot find a desired attribute in their current materials, additional germplasm must be screened. The availability of a core subset which encompasses most of the genetic diversity of the crop (and its relatives) may enhance the efficiency of identifying useful source materials and, consequently, may reduce costs.

Frankel and Brown suggest that "A core collection consists of a limited set of accessions derived from an existing germplasm collection, chosen to represent the genetic spectrum in the whole collection. The core should include as much as possible of its genetic diversity."
 
The core subset may be of greatest value to the scientist when there is little or no information regarding the most probable source of a desired attribute, which is often so with resistance to a new pathogen strain or a new pest biotype. When the passport and characterization databases which include information pertaining to the desired attribute are available, candidate accessions can often be listed (e.g. , accessions from areas with low soil pH when a source of aluminum tolerance is needed); but once such a list is made, accessions included in both the core subset and the list should be screened first. Only one overall core subset is needed for each crop collection, although the core may be developed in sections by working within species or other subgroups within the crop collection.
 
Curators of genetic resources can increase operational efficiencies by maintaining enough seeds or other propagules of accessions in the core subset to meet the expected increased demand for distributions from these samples. Although each of the total crop collections will be maintained in the NPGS base and active collections, requests for screening of accessions not in the core subset normally will not be expected until appropriate accessions in the core subset have been evaluated.
 
International Agricultural Research Centers (IARCs) in the CGIAR system maintain world collections of many crops. Most IARCs have developed or are in the process of developing core subsets for crops under their mandate. Close cooperation between NPGS and the IARCs in the development of core subsets is essential.
 
 
FORMING CORE SUBSETS
 
General
 
Development of a useful core subset will vary among crops but may involve the following steps:

1. Assembling and reviewing passport data and other information to be used in establishing non-overlapping groups.

2. Assigning accessions to appropriate groups.

3. Choosing accessions for the preliminary core subset from each group.

4. Collecting data on phenotypic and genetic traits for accessions in the preliminary core subset and using multivariate analytical methods to construct clusters and dendrograms to elucidate systematic and statistical genetic relationships for further refinement of the core subset.

5. Validation and refinement.

Another step often is needed: 2a) Acquiring additional collections from under-represented or missing groups. Additional collections are frequently needed for wild and weedy crop relatives.
 
When resources are available to characterize and statistically analyze the entire crop collection for several descriptors, steps 2, 3 and 4 can be conducted simultaneously. Groups generated from statistical analyses of the data will usually be the most robust. If data for only a few descriptors were analyzed initially, additional descriptors may be measured for the preliminary core subset, and then step 4 repeated with data from all available descriptors. When resources are limited or large numbers of accessions must be characterized, steps 2, 3 and 4 will need to be completed sequentially. Choosing 1.5 to 3 three times the number of accessions desired for the final core in step 3 is desirable (e.g., choose 15 to 30% in step 3 for characterization and evaluation in step 4 if 10% will be chosen for the final core).
 
 
Grouping and Sampling

Brown recommended stratified sampling methods when establishing core subsets. Grouping begins with taxonomic affinity (e.g., species, subspecies, cytological races). Accessions within each taxon can then be assigned to strata based on ecogeographic zones and genetic characteristics (e.g., ploidy level, photoperiod response, races, etc.). Often the distribution patterns for genetic characteristics largely coincide with ecogeographic zones. Groups such as races of maize (based primarily on ear morphology) may be preferable to countries of origin for defining groups because geopolitical boundaries often are incongruent with ecogeographic niches. In other crops, country of origin (or region of adjacent countries) may be the only reasonable means for developing preliminary groups.
 
Improving the accuracy of passport information (e.g., country of origin, taxonomic identification, race, etc.) is very important for assigning accessions to appropriate groups. When some accessions lack data regarding criteria used for grouping (origin, race, etc.), a group containing these "unknown" accessions can be formed so that the core still includes a representative sample. In step 4, the statistical analysis should assign unknown samples to appropriate clusters. Inclusion of a group of elite cultivars, or parental lines, and elite breeding populations may be desirable for most crops.
 
Brown recommended that the core subset comprise about 10% of the crop collection, but this may vary from 5% for very large collections to 50% or more on very small collections, with about 3000 suggested as a maximum number.
 
Proportional sampling (log proportional, square root, etc.) within each group may provide a more representative sample of the total genetic diversity in the core subset than would a completely random sampling from the crop collection. Multiplication by an appropriate factor to obtain the total number of core accessions equal to the desired percentage of the crop collection may be necessary. Heavier weighing of strata for primary and secondary centers of genetic diversity may have merit.
 
Once the number needed from each group has been determined, accessions for the core subset are usually chosen randomly within each group. However, some curators are choosing accessions with more desirable agronomic traits within each group. Pedigree information can be used to ensure that maximum genetic diversity is included within the group of improved cultivars or parental lines.
 
As soon as the preliminary core subset has been developed, this information should be recorded in GRIN and utilized.
 
 
REFINING THE CORE SUBSET
 
With the reduced number of accessions in the preliminary core subset, characterizations and evaluations can be conducted to obtain data needed for statistical analyses to measure genetic divergence and diversity within the core. Phenotypic and genotypic data have been used. Heritability, degree of polymorphism, and distribution of genetic factors among chromosomes affect reliability of groupings. A minimum of 10 descriptors is suggested, and more than 15 would be better.
 
Assigning heavier weights to genetic descriptors and highly heritable phenotypic traits may improve clustering. After clustering, confirmation by ordination in 2 or 3 dimensions from a principal component analysis has merit. Use of molecular characterizations, such as microsatellite, RAPD, and RFLP markers, is encouraged whenever feasible.
 
Clusters generated by multivariate analyses may provide a better understanding of patterns of genetic divergence and diversity and will often identify ecogeographic regions that have not been adequately sampled, especially when the origin of each accession is plotted geographically. This information may be valuable in planning future acquisitions.
 
When statistical analyses reveal close similarities among certain core accessions, one accession may be retained in the core and other very similar accessions may be replaced by new samples (usually from the same group, but sometimes from an under-represented group). For the larger clusters, statistical analyses can often assist with reducing the number of accessions in the core without decreasing the genetic diversity included in the core subset.
 
The core subset should remain dynamic, with accession additions, deletions, and substitutions as additional pertinent information becomes available and as new accessions are acquired. Nevertheless, with time, changes to the core should decrease in frequency and magnitude.
 

 
PLANT GERMPLASM OPERATING COMMITTEE
 
Core Subset Sub-Committee

S. Eberhart, Chair⁽¹⁾
R. Johnson
S. Kresovich
W. Lamboy
R. Schnell
C. Sperling
M. Widrlechner

August 22, 1994
 

Basigalup, D.H., D.K. Barnes, and R.E. Stucker. 1995. Development of a core collection for perennial Medicago plant introductions. Crop Sci. 35:1163-1168.

Beer, S.C., J. Goffreda, T.D. Phillips, J.P. Murphy, and M.E. Sorrells. 1993. Assessment of genetic variation in Avena sterilis using morphological traits, isozymes, and RFLPs. Crop Sci. 33:1386-1393.

Brown, A.H.D. 1989. The case for core collections. In Brown, A.H.D., Frankel, O.H., Marshall, D.R. and Williams, J.T. (eds.), The Use of Plant Genetic Resources, pp. 136-156. Cambridge Univ. Press, Cambridge.

Brown, A.H.D. 1989. Core collections: A practical approach to genetic resources management. Genome 31:818-824.

Brown, A.H.D. 1992. Genetic variation and resources in cultivated barley and wild Hordeum. Barley Genetics 6:669-682.

Brown, A.H.D. 1995. The core collection at the crossroads. In T. Hodgkin, et al. (eds.) Core collections of plant genetic resources. John Wiley and Sons, Chichester, U.K.

Brown, A.H.D., J.P. Grace, and S.S. Speer. 1987. Designation of a "core" collection of perennial Glycine. Soybean Genetics Newsletter 14:59-70.

Chang, T.T. 1989. The case for large collections. In Brown, A.H.D., Frankel, O.H., Marshall, D. R. and Williams, J.T. (eds.), The Use of Plant Genetic Resources, pp. 123-135. Cambridge Univ. Press, Cambridge.

Crossa, J., S. Taba, S. A. Eberhart, P. Bretting, and R. Vencousky. 1994. Practical considerations for maintaining germplasm in maize. Theor. Appl. Genet. 89:89-95.

Dahlberg, J.A., X. Zhang, G. E. Hart, and J. E. Mullet. (Submitted) Comparison of Clustering Methods of RAPD Marker Analysis and Statistical Clusters Based on Seed Imaging Techniques for use in Development of the Sorghum Core Collection. Crop Sci. (In review)

Diwan, N., G. R. Bauchan, and M.S. McIntosh. 1994. A core collection for the United States annual Medicago germplasm collection. Crop Sci. 34:279-285.

Erskine, W., and Muehlbauer, F.J. 1991. Allozyme and morphological variability, outcrossing rate and core collection formation in lentil germplasm. Theor. Appl. Genet. 83:119-125.

Frankel, O.H. 1984. Genetic perspectives of germplasm conservation. In Arber, W., Llimensee, K., Peacock, W.J. and Starlinger, P. (eds.), Genetic Manipulation: Impact on Man and Society, pp. 161-170. Cambridge Univ. Press., Cambridge.

Frankel, O.H., and A.H.D. Brown. 1984. Current plant genetic resources - a critical appraisal. In Genetics: New Frontiers, Vol. IV, pp. 1-11. Oxford & IBH Publ. Co., New Dehli.

Frankel, O.H., and M.E. Soulé. 1981. Conservation and Evolution. Cambridge Univ. Press, Cambridge.

Franco, J., J. Crossa, J. Diaz, S. Taba, J. Villasenor, and S.A. Eberhart. 1997. A Sequential Clustering Strategy for Classifying Gene Bank Accessions. Crop Science 37:1656-1662.

Franco, J., J. Crossa, J. Villasenor, S. Taba and S.A. Eberhart. 1997. Classifying Mexican maize accessions using hierarchical and density search methods. Crop Science 37: 972-980.

Franco, J., J. Crossa, J. Villasenor, S. Taba and S.A. Eberhart. 1998. Classifying genetic resources using categorical and continuous variables. Crop Sci. 38: (In press).

Franco, J., J. Crossa, J. Villasenor, A. Castillo, S. Taba and S.A. Eberhart. 1999. Resources in multiple environments. Crop Sci. (In press).

Harlan, J.R. 1972. Genetic resources in sorghum. In Rao, NGP, and L.R. House (eds.) Sorghum in the seventies, pp. 1-13. Oxford and IBH Publ. Co., New Delhi.

Hodgkin, T., A.H.D. Brown, Th.J.L. van Hintum and E.A.V. Morales (eds.). 1995. Core collections of plant genetic resources. John Wiley and Sons, Chichester, U.K.

Holbrook, C.C., W.F. Anderson, and R.N. Pittman. 1993. Selection of a core collection from the U.S. germplasm collection of peanut. Crop Sci. 33:859-861.

Knüpffer, H., and Th.J.L. van Hintum, 1995. The Barley Core Collection - an international effort. In T. Hodgkin, et al. (eds.) Core collections of plant genetic resources. John Wiley and Sons, Chichester, U.K.

Kouame, C.N., and K.H. Quesenberry. 1993. Cluster analysis of a world collection of red clover germplasm. Genet. Res. and Crop Evol. 40:39-47.

Marshall, D.R. and A.H.D. Brown. 1975. Optimum sampling strategies in genetic conservation. In Frankel, O.H. and J.G. Hawkes (eds.), Genetic Resources for Today and Tomorrow, pp. 53-80. Cambridge Univ. Press, Cambridge.

National Research Council 1991. Managing global genetic resources: the U.S. National Plant Germplasm System. National Academy Press, Washington, D.C.

National Research Council. 1993. Managing global genetic resources: agricultural crop issues and policies. National Academy Press, Washington, D.C.

Nei, M., and W.H. Li. 1979. Mathematical models for studying genetic variation in terms of restriction endonucleases. PNAS USA 76:5269-5273.

Peeters, J.P., and J.A. Martinelli. 1989. Hierarchical cluster analysis as a tool to manage variation in germplasm collections. Theor. Appl. Genet. 78:42-48.

Perry, M.C., M.S. McIntosh, and A.K. Stoner. 1991. Geographical patterns of variation in the USDA soybean germplasm collection. II. Allozyme frequencies. Crop Sci. 31:1356-1360.

Schoen, D.J., and A.H.D. Brown. 1995. Maximizing genetic diversity in core collections of wild crop relatives. In T. Hodgkin, et al. (eds.) Core collections of plant genetic resources. John Wiley and Sons, Chichester, U.K.

Smith, O.S., and J.S.C. Smith. 1992. Measurement of genetic diversity among maize hybrids; a comparison of isozymic, RFLP, pedigree, and heterosis data. Maydica 37:53-60.

Sneath, P.H.A., and R.R. Sokal. 1973. Numeric taxonomy. W.F. Freeman and Company, San Francisco, CA.

Spagnoletti Zeuli, P.L., and C.O. Qualset. 1993. Evaluation of five strategies for obtaining a core subset from a large genetic resource collection of durum wheat. Theor. Appl. Genet. 87:295-304.

Spagnoletti Zeuli, P.L., and C.O. Qualset. 1987. Geographical diversity for quantitative spike characters in a world collection of durum wheat. Crop Sci. 27:235-241.

Taba, S., J. Diaz, F. Pineda E., J. Franco, and J. Crossa. 1998. Pattern of Phenotypic Diversity of the Caribbean Maize Accessions. Crop Sci. 38: (In press).

Tohme, J., D. O. Gonzales, S. Beebe, and M. C. Duque. 1996. AFLP analysis of gene pools of a wild bean core collection. Crop Sci. 36:1375-1384.

Tohme, J., P. Jones, S. Beebe, and M. Iwanaga. 1994. The combined use of agroecological and characterization data to establish the CIAT Phaseolus vulgaris core collection. In T. Hodgkin, et al. (eds.) Core collections of plant genetic resources. John Wiley & Sons, Chichester, U.K.

Vaughan, D.A. 1991. Choosing rice germplasm for evaluation. Euphytica 54:147-154.

Williams, W.T. 1971. Principles of clustering, Ann. Rev. Ecol. Syste. 2:303-326.
 

7/15/98

1.  ¹Director, ¹Director, USDA, ARS, National Seed Storage Laboratory, 1111 S. Mason St., Fort Collins, CO 80521-4500.

{Status of Core Subsets}

SUBJECT: Status of NPGS Core Subsets

TO: PGOC Committee

FROM: Core Subset Sub-Committee
 

As was indicated in the 1997 Core Subset Committee report, the value of developing a core subset for each of the more than 50 species of seed propagated crops and more than 14 vegetatively propagated crops that are widely grown in the U.S. was indicated in the National Plant Germplasm System General Guidelines for Developing Core Subsets and the attached references. As shown in the attached Status Report for 1998, a core subset has been designated for each of 23 seed propagated crops and 15 vegetatively propagated crops. For the many crops with no core subset designated, the 1997 PGOC emphasized that "... the curators need to be the driving force behind their establishment and should get assistance from the CGCs." There are 20 seed crops with more than 1,500 accessions that do not yet have a core subset designated. For each location the number of crops with over 1,500 accessions with no core subset and with a core designated are as follows: Aberdeen 4/2; Ames 6/1; College Station 1/0; Geneva 2/0; Griffin 3/6; Oxford 1/0; Pullman 3/13;Sturgeon Bay 0/1; and Urbana 1/0. The vegetative crops are as follows: Corvallis 0/12; Davis 3/0; Geneva 0/2; Riverside 1/0; Sturgeon Bay 0/1.

Only the preliminary core subsets for Malus and Vitus at Geneva and Lotus and Phaseolus at Pullman have been refined with molecular methodologies. As reported last year, most accessions designated in the core subsets are available for distribution and most of them are at least partly characterized, but more work needs to be done on most of the core subsets.

Taba and staff at CIMMYT have used the very useful multivariate clustering procedures utilizing continuous and discrete variables to improve the resulting clusters or groups developed by Crossa and staff (Franco, et al., 1997) to develop a preliminary maize core subset (20%) for the Latin American landrace collections evaluated in LAMP. Plans are being developed to evaluate additional accessions in LAMP II and to evaluate the preliminary core subset to refine it to 10% of the total collection.

As indicated in the 1997 report, if we can represent 50 to 70% of a crop's diversity (depending on the quality of the procedures for designating the core) in about 10 % of the accessions designated as the core subset, curators can put priorities for regeneration and characterization on the 10 to 15% in the core subset to be sure adequate seed is available and that users will have ready access to nearly 70% of the diversity for that crop. The cost of designating the core with geography (15 to 30%) and refining to 8 or 10% with multivariate analyses of characterization data and/or with molecular data on this 15 to 30% can be reasonable.
 

Core Subset Sub-Committee Members:

S. Eberhart , Chair                         H. Bockelman
N. Garvey                                         R. Johnson
S. Kresovich                                   W. Lamboy
R. Schnell                                        M. Widrlechner