GRIMS glossary
From ICISWiki
GRIMS main > GRIMS documentation
Contents |
GRIMS Concepts and Terminology
This page contains the definitions of concepts and terminology that are used in ICIS and GRIMS. They are intended to be generically applicable to collections of plant genetic resources managed ex situ as seed samples, although they have been developed largely in the context of rice at IRRI. Feel free to make changes, add new terms, or use the discussion page for questions, suggestions, etc.
Disclaimer: These definitions are in a draft form. Once they have been
widely discussed and accepted there will be a message saying so.
Underlying germplasm model
Germplasm sample
A germplasm sample can be thought of as a packet of seeds or other propagating material that is managed separately from other such packets of seeds in a collection, in the sense that germplasm specialists would not want to mix them. A new germplasm sample is created from an existing sample when any one of the following events occurs:
- A fresh lot of seed or other propagules is harvested (e.g. seeds from a plot, farmer’s field or natural ecosystem). Each cycle of propagation involves a certain amount or probability of genetic change by mutation, drift, selection (deliberate or unintentional) and/or introgression. Even pure lines may change genetically during propagation, and therefore successive generations of propagation should not be mixed; instead they must be treated as distinct samples.
- A heterogeneous sample is split into 2 or more genetically distinct components
- A sample or random sub-sample is transferred to management by a different Organizational Unit (e.g. transferring between genebanks or breeding programmes; collecting samples from a market place of farm store)
- A component of a plant (such as DNA, protein, etc.) is extracted for analysis
Germplasm creation method
The concept of Germplasm creation method (abbreviated in ICIS documentation and applications to "Germplasm method") is biologically motivated. It describes the method by which the germplasm identified by a GID was created from its most recent known progenitor, which may be the immediate parents or an unknown number of generations before. Based on the intended change of genetic diversity from parent to offspring, we distinguish three classes of Germplasm creation methods:
- Generative germplasm creation methods (intend to increase genetic diversity, e.g. cross)
- Derivative germplasm creation methods (intend to reduce genetic diversity, e.g. selection of a particular variant from a heterogeneous population)
- Laboratory Derivative methods that isolate a component of a plant for analysis or extract/modify molecular components from a tissue sample (such as DNA, protein, etc.) may also be treated as a subvariant of Derivative germplasm creation method
- Maintenance germplasm creation methods (intend to maintain genetic diversity, e.g. rejuvenation)
Germplasm Neighbourhood
The concept of neighbourhood is used to describe sets of germplasm samples that somehow "belong together" – for biological, experimental, legal, or managerial purposes. For biological neighbourhoods we define derivative, maintenance, and management neighbourhoods to identify the type of biological relationship between germplasm samples.
Derivative neighbourhood
A germplasm sample that is produced through a derivative germplasm creation method or a maintenance germplasm creation method has one immediate progenitor, referred to in ICIS applications as the seed source (but be careful not to confuse this usage of seed source with other common usages; e.g. in IRRI, to the Seed Health Unit the seed source is the scientist who gave the seed to the SHU, and to the Genetic Resources Centre the seed source is the institute that sent the seed to IRRI; also distinguish from group source below). If we trace back the genealogy of germplasm samples produced only through derivative or maintenance germplasm creation methods, we will end up at a germplasm sample that either has unknown progenitors (such as a landrace or a breeding line of unknown provenance), or is the product of a generative germplasm creation method. We consider this germplasm sample as the root of a derivative neighbourhood, also called the group source. All germplasm samples produced from this root sample through either derivative or maintenance germplasm creation methods form a derivative neighbourhood, all derived from one common progenitor. It should be noted that a derivative neighbourhood may contain a number of maintenance neighbourhoods, and the germplasm samples in it may be held by more than one institution. Starting from a germplasm sample anywhere in the derivative neighbourhood and tracing back and forth between samples linked by derivative and maintenance methods will eventually cover the whole derivative neighbourhood. However nested subsets of this neighbourhood derived by limiting the number of steps forward or backward in genealogy provide interesting neighbourhoods and we call these 'n step derivative neighbourhoods' where n is the limit on the number of steps.
Maintenance neighbourhood
All germplasm samples produced from a common progenitor sample through maintenance germplasm creation methods form a maintenance neighbourhood. In a maintenance neighbourhood, all germplasm samples are produced from one common progenitor with the aim of maintaining the genetic diversity of that common progenitor. A common example is a modern released cultivar. Breeders and seed companies try to maintain the genetic identity of a cultivar fixed, despite repeated cycles of propagating new samples of the cultivar and distribution to different parts of the target market area. All samples of one such variety are linked in a single maintenance neighbourhood. (By contrast, a landrace or traditional cultivar is not a maintenance neighbourhood: such cultivars have been jointly developed by farmers, often from many communities over many years. Extant samples of a single landrace can be genetically distinct and only distantly related, through many generations of hybridisation, introgression and selection. Commonly each sample of a landrace is heterogeneous and farmers commonly make their own selections each year, as they seek to improve the landrace rather than maintain its identity)
Management neighbourhood
A management neighbourhood is a special kind, and may be a subset, of a maintenance neighbourhood. In addition to being in the same maintenance neighbourhood, all the germplasm samples in a management neighbourhood are under the responsibility [sic; note responsibility NOT necessarily ownership] of a single institution. For example, the stocks of a modern released cultivar maintained by one seed company would form one management neighbourhood of the variety’s maintenance neighbourhood.
Management of genetic resources in genebanks
Accession
In a genebank, accessions are the key genetic entities. An accession is the genetic unit of maintenance, responsibility, exchange, distribution, and intellectual property. Genebanks uniquely identify accessions within their collection through Accession IDs (numbers, names). Recipients of germplasm from a genebank identify the germplasm they need by Accession ID and its characteristics, and are expected to use the Accession ID to identify the germplasm in their publications.
Just as a seed company seeks to maintain the genetic identity of an improved variety, so a genebank seeks to maintain the genetic identity of an accession despite repeated cycles of propagation. Information on the origins of an accession and on any conditions governing its use remain unchanged over cycles of propagation. As part of the process of maintaining the genetic identity of an accession, genebanks need to track the cycles of propagation, and even to divide each germplasm sample into sub-samples needed for different elements of genebank operations. However, the primary aim of tracking these smaller units is to ensure effective conservation and use of the accession.
An accession is thus a set of Germplasm samples in a management neighbourhood. It consists of a Founding sample and all germplasm samples produced from the Founding sample through Germplasm Maintenance methods under the management of the genebank.
Founding sample
A founding sample is a sample of seeds (or other propagating material) that enters a genebank through a donation, a collecting activity, or some other activity that has resulted in a possibly unique germplasm entity that may be worthwhile preserving. In some genebanks, an accession ID is assigned to a founding sample immediately upon receipt. In others, the founding sample is first subjected to initial checks and processing, and it will be accessed into the collection, with the assignation of an accession ID, only after successful completion of the initial processing. This typically includes testing for health, viability, uniformity, and possible duplication in the collection, and if necessary it may include undertaking an initial rejuvenation or multiplication of seed. In the case of founding samples that are mixtures, the sample may be divided into its components with a separate accession ID assigned to each component. The founding sample is genetically the “best” possible sample of an accession, as it the least changed by drift, selection and contamination. If original seed remain after completion of the initial tests, it is desirable to keep them under optimal conditions in long term storage – not for routine use, but for checking on questions that may arise about unintentional drift, selection and contamination. However, in many cases, no viable seeds of the founding sample are available, but only seeds from subsequent generations produced through germplasm maintenance methods, such as multiplication or rejuvenation.
Incoming sample
A synonym for Founding sample
Registered sample
see Accession
Unregistered sample
see Founding sample
Representation of germplasm and germplasm relationships in ICIS
Germplasm ID (GID)
The Germplasm ID (GID) is a unique identifier for a Germplasm record within an ICIS database. GIDs are integer values that are automatically generated by the database when new germplasm records are created. GIDs are simple identifiers without any intrinsic information content, except that GIDs in a local ICIS database have negative values, whereas GIDs in a central ICIS database have positive values. Their primary use is for rigorous internal identification of germplasm within the database, not for external publication.'
A GID may identify:
- A Germplasm sample currently or previously in a genebank, breeder’s or other collection
- A Germplasm Neighbourhood currently or previously in a collection. (For example, each rice variety in INGER is managed as a single GID that encompasses all past, current and future samples of the neighbourhood; contrast the same varieties in the genebank, where each generation is identified using a different GID)
- A notional Germplasm sample that has never existed in a collection – for example, a crop in a farmer’s field or a bin in a market store from which a sample was collected.
- A notional Germplasm neighbourhood that has never existed in a collection – for example a released variety that may have been grown many times by many farmers in many places but represented in an ICIS database by only one GID.
- A laboratory derivative product of a Germplasm sample such as a tissue sample, DNA or protein extract or similar downstream reagent of germplasm origin. This kind of GID serves as a kind of sample tracking identifier in a laboratory information management system.
GIDs of the notional samples and neighbourhoods are typically assigned for such germplasm as part of the documentation of the origins of and relationships between GIDs representing samples and neighbourhoods in a collection.
Germplasm creation method
The Germplasm creation method (of the germplasm sample if the GID represents a sample; or of the root sample of the neighbourhood if the GID represents a neighbourhood) is stored in the GERMPLSM table of the ICIS GMS database. Fields in this table
- specify the date of germplasm creation
- link the GID to the location of the germplasm creation
- link the GID to its progenitor sample(s) (via GIDs)
- specify a Germplasm creation method code that defines the (biological) nature of this relationship. The Germplasm creation methods are user defined and stored in the METHODS table of the GMS database, with a central repository kept in the central database, e.g. IRIS in case of rice. The management of Germplasm creation method definitions are currently being generalized into a proper controlled vocabulary and ontology within the context of the Generation Challenge Programme
Derivative neighbourhood
Every GID in a derivative neighbourhood has a pointer (GPID1) to the GID of the group source as well as a pointer (GPID2) to the GID of the seed source (ICIS applications commonly abbreviate group source and seed source respectively to group and source). That is, for any germplasm sample in a derivative neighbourhood we explicitly use the GID of the group source (root germplasm sample) as an indicator of membership in that derivative neighbourhood. This is useful, since in many derivative neighbourhoods there is no unbroken line of genealogy information, but the root germplasm sample is known (e.g. a cross). This is typically the case in the selection of lines from an initial cross, where the information on intermediate generations has not been recorded.
Maintenance neighbourhood
All germplasm produced from a common progenitor sample through maintenance germplasm creation methods form a maintenance neighbourhood. In a maintenance neighbourhood, all germplasm samples are produced from one common progenitor with the aim of maintaining the genetic diversity of that common progenitor. Unlike derivative neighbourhoods, where germplasm samples have an explicit link to the common progenitor sample (group source), maintenance neighbourhoods can be identified only by tracing lines of genealogy and germplasm creation method records. The germplasm samples in a maintenance neighbourhood may be held by more than one institution.
Management neighbourhood / Accession
Effective management of an accession depends on maintaining 1 GID for each germplasm sample in the accession. For any GID representing a germplasm sample in a Management neighbourhood we explicitly use the GID of the founding sample as an indicator of membership in that management neighbourhood. All germplasm samples that belong to an accession have an associated GID (MGID) that points to the record of the founding germplasm. This is necessary, since the passport data are connected to the founding germplasm, but need to be linked to all germplasm sample records in the accession.
However, in many cases, particularly for accessions received from other collections, a collection manager may have no information on the individual samples that comprise the management neighbourhood of the donor’s accession. In these cases, one GID is used to represent the entire management neighbourhood of the accession, not just its founding sample
