Skip to main content

Minimum data requirements

While the Multi-Crop Passport Descriptors (MCPD) v2.1 standard lists 28 main descriptors, Genesys doesn't require all of them.

We treat the MCPD descriptors in three tiers:

  • Mandatory descriptors: the absolute minimum needed to publish an accession. If these fields are not populated, Genesys cannot publish the record.
  • Strongly recommended descriptors: fields that dramatically improve discoverability and cross-collection comparison. Genesys does not reject records that lack these, but it will flag them, and your PDCI score will be low.
  • Optional descriptors: useful where you have the information, but not blocking. Many of these fields apply only to certain types of accessions (e.g., breeding institute information is only relevant for breeding material).

In this section, we'll walk you through the mandatory fields in detail, as these account for most rejection errors during upload.

Mandatory fields​

Descriptor nameField codeWhat it holds
Institute codeINSTCODEYour FAO WIEWS institute code (the genebank holding the accession)
Accession numberACCENUMBYour unique identifier for this accession within your collection
GenusGENUSThe genus the plant belongs to

Together, these three fields create a unique reference for the accession in Genesys when DOI is not available.

INSTCODE: The FAO WIEWS institute code​

The DPA's required information section covers how to find or obtain your INSTCODE. This is the globally unique identifier assigned by FAO WIEWS, typically a three-letter country code plus a number, e.g., LBN020 or BRA003.

Here are some common pitfalls:

  • Wrong INSTCODE used: for institutes with multiple WIEWS codes (e.g., a parent institute and regional branches), make sure each accession is tied to the WIEWS code of the genebank that actually holds it
  • INSTCODE missing: accessions cannot be published without one. If your institute does not yet have a WIEWS code, obtain one before uploading
  • Old or obsolete INSTCODE: if your institute has been re-registered with FAO under a new code, use the current code throughout your data

ACCENUMB: The accession number​

This is your unique identifier for an accession within your collection. It is the value you use internally, on labels, in your local database, and in your records.

Here are some common pitfalls:

  • Duplicate ACCENUMB values within the same INSTCODE: every accession number must be unique within the institute. If you have historically reused numbers (e.g., after deaccessioning a sample), the duplicates must be resolved before upload
  • Special characters that break processing: avoid characters that could be misinterpreted by spreadsheet or database systems, e.g., commas, tabs, line breaks, or quotation marks within the accession number itself
  • Leading-zero stripping: if your accession numbers contain leading zeros (e.g., 00012345), make sure your spreadsheet preserves them as text rather than converting them to numbers (see preparing MCPD-compliant data section on Excel formatting)

GENUS: The genus name​

The genus name should be correctly spelled and capitalized.

Here are some common pitfalls:

  • Spelling errors: even small errors prevent matching across collections. It's Triticum not Triticom, and Phaseolus not Phaesolus.
  • Combined genus + species in one field: Triticum aestivum belongs in two separate fields: GENUS = Triticum, SPECIES = aestivum. Combining them breaks the validation.
  • All-uppercase or all-lowercase: genus names are capitalized by convention: it's Triticum, not TRITICUM or triticum.

Beyond the mandatory fields, the following descriptors significantly improve how people can use your records. Without them, your accessions are publishable but much harder to find, compare or interpret.

Taxonomy details​

  • SPECIES: the species epithet (e.g., aestivum). Without this, your accession is only identified to genus level, which severely limits search precision
  • SPAUTHOR: the taxonomic authority for the species name (e.g., L. for Linnaeus). This is useful for taxonomic precision, but is a lower priority than species epithet
  • SUBTAXA / SUBTAUTHOR: subspecies, botanical variety or form, where applicable

Origin and identifiers​

  • COLLNUMB: the collector's original identifier. This is essential for identifying duplicates held in different collections. If accession X in your collection and accession Y in another collection share a collecting number, they probably trace back to the same original sample
  • ACCENAME: accession name(s), variety name or local landrace or heirloom name. Researchers and breeders often search by name
  • ORIGCTY: country of origin (three-letter ISO 3166-1 code). This is one of the most common search filters

Status​

  • SAMPSTAT: biological status of the accession (wild, landrace, breeding material, cultivar, etc.: see the code list in preparing MCPD-compliant data). This is one of the highest-impact fields for user search
  • MLSSTAT: Multilateral System status under the ITPGRFA. This is important for users navigating availability

Collecting context​

If the accession was collected (as opposed to bred or donated):

Optional fields​

The remaining MCPD descriptors are optional because they only apply to certain types of accessions. Examples:

  • BREDCODE / BREDNAME, only applies to breeding material; not relevant for wild or landrace accessions
  • DONORCODE / DONORNUMB, applies only to accessions received from a donor institute
  • DUPLSITE, applies only to accessions that have safety duplicates elsewhere
  • REMARKS, useful for any free-text notes that do not fit elsewhere

Populating these where they apply increases your PDCI score (data curation and validation) and improves the usefulness of your records, but their absence does not prevent publication.

A reasonable minimum target​

The three-field absolute minimum lets you publish, but it produces records that are not particularly useful to anyone. A better target for new Data Providers, which we encourage aiming for, is the following set (populated where the information exists):

  • INSTCODE, ACCENUMB, GENUS (mandatory)
  • SPECIES, COLLNUMB, ACCENAME (strong identifiers)
  • ORIGCTY, SAMPSTAT, MLSSTAT (strong filters)
  • For collected accessions: COLLSITE, DECLATITUDE, DECLONGITUDE, COLLDATE
  • STORAGE: type of germplasm storage, which helps users to know what sort of plant parts they are going to receive if they request the accession

This is 10-13 fields, all of which most genebanks already track in their internal records.

As we discuss in data curation and validation, our principle is: publish what you have now, and improve it over time. Do not delay publication for years trying to populate every field fully. Get the mandatory and strongly recommended fields right, publish, and iterate.

Common pitfalls and rejection errors at upload​

Aside from mandatory-field checks, several recurring errors can cause uploads to fail. The Validator Tool catches most of these before you attempt an upload. Running your file through it first is much better than trying to diagnose errors after a rejected upload.

  • Duplicate ACCENUMB within the same INSTCODE: every accession number must be unique
  • GENUS missing or unrecognized: taxonomic validation runs against a reference list; misspellings are flagged
  • Invalid country code: e.g., LB instead of LBN or Lebanon instead of LBN
  • Invalid INSTCODE: institute codes are validated against the live FAO WIEWS list; an INSTCODE that does not exist (or has been retired) will fail
  • Date format errors: dates not in YYYYMMDD format
  • Invalid controlled-vocabulary value: e.g., landrace instead of 300 in SAMPSTAT, or a code not in the defined list
  • SAMPSTAT 999 without a REMARKS prefix: using the "Other" code without explaining in remarks

The data curation and validation section covers the Validator Tool's output in detail.