Preparing MCPD-compliant data

In this section, we provide a practical reference to help you get your passport data into the right shape before upload. We cover file structure, formatting rules for common field types, controlled vocabularies, coordinates, and the most frequent mistakes new providers make.

The section on minimum data requirements explains which descriptors we treat as mandatory. This section assumes you've already decided which descriptors you want to publish and now need to format them correctly.

tip

The Crop Trust has a multi-part Passport Data Preparation webinar series that demonstrates these steps live and is worth watching alongside this written reference. Links to the relevant sessions are listed at the end of this section.

File structure: one row per accession

We accept passport data in tabular form, typically as a Microsoft Excel file (.xlsx) or a comma-separated values file (.csv). The structure is simple:

• One row per accession. Each accession in your collection is one row. If you have 5,000 accessions, you have 5,000 data rows.
• One column per MCPD field. Each column corresponds to one MCPD descriptor.
• A header row at the top. The first row contains the column headers, and these headers must use the exact MCPD field codes (described below), not free-form labels.

A small example with five accessions and a subset of fields:

INSTCODE	ACCENUMB	GENUS	SPECIES	ORIGCTY	ACQDATE	SAMPSTAT
LBN020	LB-1001	Triticum	aestivum	LBN	19850600	300
LBN020	LB-1002	Triticum	durum	LBN	19850600	300
LBN020	LB-1003	Hordeum	vulgare	SYR	19870415	100
LBN020	LB-1004	Lens	culinaris	LBN	19920000	300
LBN020	LB-1005	Cicer	arietinum	TUR	20010512	500

The Genesys Uploader Tool and the validator at validator.genesys-pgr.org both expect this format.

Use the exact MCPD field codes as column headers

The MCPD assigns a short code (called the "suggested fieldname") to each descriptor. We use these codes to map your columns to our internal fields. Use the exact codes: case-sensitive, no spaces, and no extra punctuation.

Common field codes you will use:

Descriptor name	Field code	What it holds
Institute code	INSTCODE	Your FAO WIEWS institute code (e.g. LBN020)
Accession number	ACCENUMB	Your unique identifier for the accession
Collecting number	COLLNUMB	The collector's original identifier
Collecting institute code	COLLCODE	FAO WIEWS code of the institute that collected the sample
Genus	GENUS	Botanical genus name
Species	SPECIES	Specific epithet
Species authority	SPAUTHOR	Taxonomic authority for the species name
Subtaxon	SUBTAXA	Subspecies, variety or form
Subtaxon authority	SUBTAUTHOR	Authority for the Subtaxon name
Common crop name	CROPNAME	Common name (e.g. "Bread wheat")
Accession name	ACCENAME	Variety or local name(s) given to the accession
Acquisition date	ACQDATE	Date your institute acquired the sample, as YYYYMMDD
Country of origin	ORIGCTY	Three-letter ISO 3166-1 country code
Location of collecting site	COLLSITE	Free-text description of the location
Latitude (decimal degrees)	DECLATITUDE	Latitude in decimal degrees
Longitude (decimal degrees)	DECLONGITUDE	Longitude in decimal degrees
Coordinate uncertainty	COORDUNCERT	Uncertainty in meters
Georeferencing method	GEOREFMETH	How the coordinate was derived
Elevation	ELEVATION	Elevation in meters above sea level
Collecting date	COLLDATE	Date the sample was collected, as YYYYMMDD
Breeding institute code	BREDCODE	FAO WIEWS code of the breeding institute
Biological status	SAMPSTAT	Three-digit code (see "Controlled vocabularies" below)
Donor institute code	DONORCODE	FAO WIEWS code of the donor institute
Donor accession number	DONORNUMB	Donor's identifier for the accession
Other identifiers	OTHERNUMB	Other identifiers associated with the accession
Safety duplication location	DUPLSITE	FAO WIEWS code(s) of the genebank(s) holding safety duplicates
Type of germplasm storage	STORAGE	One or more single-digit codes (see below)
MLS status	MLSSTAT	Multilateral System status under the International Treaty
Remarks	REMARKS	Free-text remarks

headers

Headers must match these codes exactly.
instcode, INSTITUTECODE, Institute Code, and INSTCODE (with a trailing space) will all fail.

Genesys extensions to the MCPD standard

The MCPD standard defines the descriptors that make up passport data. Genesys defines a small number of extensions to that standard, which are additional fields the platform recognizes for specific features. These extensions can be included in your upload file alongside the standard MCPD fields.

Five extensions apply to accession records:

Field code	Type	Purpose
ACCEURL	URL	Link to your institutional record for the accession
AVAILABLE	boolean	Whether the accession can be requested through Genesys
HISTORIC	boolean	Whether the accession is no longer held by your genebank
CURATION	enum (FULL, PARTIAL, ARCHIVED, HISTORICAL)	Fine-grained curation state of the record
UUID	UUID	Stable globally unique identifier for the record

ACCEURL: Link to your institutional record

ACCEURL is a URL pointing to your institutional record for the accession: for example, the accession's page on your genebank's own website or in your local information system. Populating this field lets users click through directly to your authoritative record from Genesys.

This is particularly useful if your institute maintains its own detailed information system that goes beyond what MCPD passport data captures, for example, extensive characterization data, historical images, or curation notes not shared through Genesys.

UUID: Stable identifier for the record

UUID records a Universally Unique Identifier for the accession record. Genesys automatically generates a UUID for each record it stores, so most providers don't need to populate this field in their upload files.

This UUID serves as the primary internal identifier for the accession record when a PUID (DOI) is not available. If your institute already assigns UUIDs to accessions in your own information system and you want us to preserve that specific identifier rather than generating a new one, you can populate this field.

UUID vs PUID

UUID is distinct from the MCPD PUID field (Persistent Unique Identifier).
In Genesys, the only accepted form for the PUID field is a DOI. Any other identifier forms (such as UUID or LSID) are ignored if placed in the PUID field. If you have a UUID for the record, use the UUID extension instead.

AVAILABLE, HISTORIC and CURATION: referenced elsewhere

Three of the extensions relate to workflows beyond initial data preparation. We cover these in more detail in later modules:

• AVAILABLE: controls whether users can request material for this accession through Genesys. See requests for material for the request workflow and how availability interacts with your genebank-level opt-in.
• HISTORIC: marks an accession as no longer held by your genebank. Historical accessions remain visible for discovery and citation but cannot be requested. See frequency of data updates for the mechanics of marking accessions as historical.
• CURATION: with values FULL, PARTIAL, ARCHIVED, and HISTORICAL. This records the curation state of the accession record. When both CURATION and HISTORIC are set, CURATION takes precedence. See frequency of data updates for how to use it.

Common formatting rules

The MCPD defines a small set of formatting rules that apply across many descriptors. These are the most common source of upload errors.

Multiple values: separate with a semicolon, no space

When a field accepts multiple values (such as multiple accession names), separate them with a semicolon and no space between values:

• Correct: Symphony;Emma;Songino
• Wrong: Symphony; Emma; Songino (extra spaces)
• Wrong: Symphony, Emma, Songino (wrong separator)
• Wrong: Symphony / Emma / Songino (wrong separator)

The same rule applies to multi-value institute fields (for example, when more than one institute collected a sample) and to lists of safety-duplication locations.

Missing values: leave the cell empty

If you don't have data for a field, please leave the cell empty. Do not write "N/A", "unknown", "-", "missing", "?" or "n.n.", as all of these are interpreted as actual data values and trigger validation errors.

• Correct: empty cell
• Wrong: N/A
• Wrong: Unknown
• Wrong: -
• Wrong: 0 (zero is a valid numeric value, not a placeholder)

For numeric fields exchanged via database, missing values should be represented by NULL, but for spreadsheet uploads, an empty cell is correct.

Dates: YYYYMMDD format

Dates use the format YYYYMMDD (year, month, day, all numeric, no separators):

• Full date known: 19850615 (15 June 1985)
• Day unknown: 19850600 (June 1985, day unknown, note the trailing 00)
• Day and month unknown: 19850000 (sometime in 1985)
• Day and month unknown, alternative form: 1985---- (both the hyphen and the double-zero forms are equally valid in MCPD v2.1)

This format applies to both Acquisition date (ACQDATE) and Collecting date (COLLDATE).

Common mistakes to avoid:

• Wrong: 15/06/1985 or 06/15/1985 (slash separators, ambiguous order)
• Wrong: 15-Jun-1985 (month name)
• Wrong: 1985 (year only, without the trailing zeros)
• Wrong: 1985-06-15 (ISO 8601 with dashes, MCPD wants no separators)

Country codes: three-letter ISO 3166-1

Countries use the three-letter ISO 3166-1 alpha-3 codes, not the two-letter alpha-2 or country names:

• Lebanon: LBN (not LB, not Lebanon)
• Brazil: BRA (not BR, not Brazil)
• Côte d'Ivoire: CIV (not IC, not Ivory Coast)
• United States: USA (not US, not United States)

You can find the full ISO 3166-1 list from the UN Statistics Division. For accessions collected in countries that no longer exist (e.g. the former Yugoslavia or USSR), use the relevant historical code from the obsolete-codes list linked in the MCPD itself.

The Crop Trust's Passport Data Preparation webinar series Part 3 (referenced at the end of this section) demonstrates this mapping in detail, including how to handle ambiguous country names and split countries.

Institute codes: FAO WIEWS INSTCODE

Institute codes follow the FAO WIEWS institute code (INSTCODE) format: a three-letter ISO country code plus a number (e.g. LBN020, BRA003, USA022). Required Information covers how to find or obtain your INSTCODE.

If you need to record an institute that does not have a WIEWS code:

• Leave the code field (e.g. COLLCODE) empty
• Use the corresponding name and address fields (e.g. COLLNAME and COLLINSTADDRESS) to record the institute's details in free text

Please do not invent ad hoc institute codes; leaving the field empty is correct when no WIEWS code exists.

Coordinates: decimal degrees, WGS84

Latitude and longitude are recorded as decimal degrees in DECLATITUDE and DECLONGITUDE:

• Latitude: positive values north, negative south (e.g. Beirut ≈ 33.8886, Buenos Aires ≈ -34.6037)
• Longitude: positive values east, negative west (e.g. Beirut ≈ 35.4955, San Francisco ≈ -122.4194)

Use WGS84 as the reference geodetic datum. If you want to be explicit, record this in COORDDATUM as WGS84.

If your historical records contain coordinates in degrees-minutes-seconds (DMS) format, the MCPD provides parallel descriptors LATITUDE and LONGITUDE for that representation. However, our geo-data validator (Data curation and validation) only accepts the decimal format, so you must convert DMS coordinates to decimal before uploading.

For coordinate generalization when sensitive locations are involved, see What is passport data?, and remember to set COORDUNCERT to make the level of precision explicit.

Controlled vocabularies: use the codes, not free text

Several MCPD descriptors require values from a defined code list. Please use the codes as specified, rather than free-text descriptions. Three are particularly important.

SAMPSTAT: biological status of the accession

SAMPSTAT records the biological status of the accession. The code is a three-digit number organized in groups:

Code	Meaning
100	Wild
110	Natural
120	Semi-natural / wild
130	Semi-natural / sown
200	Weedy
300	Traditional cultivar / landrace
400	Breeding / research material
410	Breeder's line
411	Synthetic population
412	Hybrid
413	Founder stock / base population
414	Inbred line (parent of hybrid cultivar)
415	Segregating population
416	Clonal selection
420	Genetic stock
421	Mutant (e.g. induced mutant, TILLING population)
422	Cytogenetic stock
423	Other genetic stock (e.g. mapping populations)
500	Advanced or improved cultivar (conventional breeding)
600	GMO (by genetic engineering)
999	Other (elaborate in REMARKS field)

Use the most specific code that applies. For example, prefer 412 (Hybrid) over 400 (Breeding / research material) when you know the accession is a hybrid.

If you use 999 (Other), the REMARKS field must contain an explanation prefixed with SAMPSTAT:, for example, SAMPSTAT: F2 population from cross of parents X and Y. The PDCI validator (Data curation and validation) checks for this prefix and will flag any records that use 999 without it.

COLLSRC: collecting / acquisition source

COLLSRC records where the sample came from. The code list includes:

Code	Meaning
10	Wild habitat
11	Forest or woodland
12	Shrubland
13	Grassland
14	Desert or tundra
15	Aquatic habitat
20	Farm or cultivated habitat
30	Market or shop
40	Institute, experimental station, research organization, genebank
50	Seed company
60	Weedy, ruderal or disturbed habitat (roadside, field margin, etc.)
99	Other (elaborate in REMARKS field, prefix COLLSRC:)

STORAGE: type of germplasm storage

STORAGE records how the accession is conserved. Multiple values are allowed (semicolon-separated). The code list includes:

Code	Meaning
10	Seed collection
11	Short term
12	Medium term
13	Long term
20	Field collection
30	In vitro collection
40	Cryopreserved collection
50	DNA collection
99	Other (specify in REMARKS, prefix STORAGE:)

tip

The full code lists for these and other controlled-vocabulary descriptors are in the MCPD v2.1 reference document.
When in doubt, it's best to consult the standard.

Practical Excel guidance

Since most genebanks prepare passport data in Microsoft Excel, here are a few practical points to save you time and avoid common errors.

Use one workbook, one sheet. Do not split your data across multiple sheets or multiple files. Both the validator and the Uploader Tool expect a single tabular dataset.

Set column formatting carefully. Excel sometimes automatically converts your data in ways that can cause errors:

• For example, if you enter 19850615 in the ACQDATE field, Excel may auto-convert it to a date display. We recommend setting the column to Text format before pasting your date data to prevent Excel from converting values like 19850600.
• The INSTCODE and COLLCODE fields (e.g. LBN020) are text; if Excel treats them as numbers, leading characters can be stripped or zeros can be lost. Force these columns to Text.
• The same applies to numeric fields with leading zeros (for example, accession numbers like 0001234).

Avoid hidden whitespace. Trailing spaces in text fields are a common cause of failed matches between your data and reference lists (such as the WIEWS institute directory). Use Excel's TRIM() function before saving.

Save as .xlsx or .csv. Both are accepted, but if you choose CSV, please use UTF-8 encoding to preserve accented characters in accession names or remarks.

Validate before uploading. Even if your data looks clean, we recommend running it through validator.genesys-pgr.org before uploading. The validator catches problems that aren't always visible at a glance.

A short pre-upload checklist

Before you upload your file to Genesys, scan through this list:

• One row per accession; one column per MCPD descriptor
• Header row uses exact MCPD field codes (case-sensitive, no spaces)
• No duplicate accession numbers (ACCENUMB unique within INSTCODE)
• Dates in YYYYMMDD format throughout
• Countries in three-letter ISO 3166-1 codes
• Institute codes in FAO WIEWS INSTCODE format
• Controlled vocabulary fields (SAMPSTAT, COLLSRC, STORAGE) use defined codes only
• Coordinates in decimal degrees, with COORDUNCERT populated where coordinates are generalized
• Empty cells for missing data, no "N/A", "unknown" or placeholder strings
• File validated through validator.genesys-pgr.org with no errors before upload to production

Further learning: webinar series

The Crop Trust has a multi-part Passport Data Preparation webinar series demonstrating these steps live with worked examples. The sessions are recorded and available on the Crop Trust YouTube channel:

• Passport Data Preparation, Part 1, Common Excel data-cleaning shortcuts and functions
• Passport Data Preparation, Part 2, Mapping passport Data to MCPD
• Passport Data Preparation, Part 3, Mapping to MCPD: Other data-type formatting and data validation

If you wish to receive the data sample used in the demonstrations, contact helpdesk@genesys-pgr.org