Biota: The Biodiversity Database Manager

This document analyzes the mapping of Biota's actual tables, fields, and relationships (as shown in detail in the Biota Data Model) on the entities and relationships of the ASC Information Model for Biological Collections (Association of Systematics Collections, Committee on Computerization and Networking, March 1993 Version). This analysis is intended to help potential and actual Biota users understand the relationships between the two models, and in the longer term, to lay the groundwork for interoperability between Biota data files and data files based on other modifications or implementations of the ASC model.

It is essential to understand that the ASC model is, in the words of the ASC document, a "'first cut' at a high-level information model for biological collections" (an entity-relationship model) not a complete design for a working database. For most entities (which translate to the tables of an actual database schema), the ASC model specifies the primary and foreign keys, but provides only representative examples of attributes (data fields). For some functionalities, such as loan management, the ASC model provides only the barest structural skeleton (the TRANSACTION entity). To quote from the introduction of the ASC document:

"The ASC Data Standards Workshop participants recognize that the information model presented in this report is not complete. The parts are neither fully fleshed out nor documented adequately.... Nevertheless, a decision to distribute the report at this time was made so that individuals and institutions currently modeling or planning to model biological collections data will have the opportunity to derive some benefit from the work done so far, and so that the model itself can be critiqued and refined by a broader group...."

"The model presented describes the high-level entities (logical and physical objects) that comprise the domain of a biological collection, including collecting activities, specimen objects and their handling, their taxonomy, locality and collector data, other objects, and the relationships among them. The model should be able to accommodate an insect collection as satisfactorily as a fish collection, botanical collection, or paleontological collection. Therefore, this conceptual information model provides an opportunity for the diverse disciplines of collections-based biologists to discuss their common activities and analogous information in a mutually comprehensible frame of reference. The model also provides the framework from which more detailed models can be developed.... There is no implication in this conceptual information model as to how it should be physically implemented."

While Biota provides the basic tools and functionality for institutional collections management, Biota is committed to the objective of filling the need for a personal data management tool for researchers. The philosophy behind Biota's design, then, is to exploit the power of a relational data structure, while keeping the model simple enough for individual users. The goal is to allow biologists, untrained in database design and management, to make full use of Biota's tools for data analysis, data export, and--especially--the import of pre-existing datasets, with only the assistance of the Biota manual. The ASC model, if fully implemented, is probably far too complex to meet these criteria, particularly the objective of making data import practical.

For these reasons, Biota's design is, in most regards, a simplification the ASC model, which indeed proved of considerable value in suggesting structures and relationships. The simplification was achieved in four ways:

(1) Some particularly complex data types represented by some groups of entities in the ASC model are not implemented in Biota's current design. These include and multiple simultaneous taxonomic classifications and paleontological data (with apologies to paleontologists).

(2) The recursive relationships in the ASC model for taxonomic classification and place-names were replaced with hierarchical structures. For most biologists, it is much easier to conceptualize both taxonomic and geographic information in this traditional design, and considerably simpler to prepare and import text files in this format than in the recursive format.

(3) Where practical, child entities in the ASC model were modeled as attributes (fields) of the parent entity (table) in Biota. For example, the ASC entity COLLECTING-METHOD (many-to-many with COLLECTING-EVENT) corresponds to the Method field of the [Collection] (collecting-event) table in Biota.

(4) In the ASC model, many entities are modeled at supertypes with subtypes (for example, GEOMETRIC-LOCALITY is the supertype for subtypes POINT, LINE, and POLYGON). Biota does not implement subtypes, as such. In some cases, the commonest ASC subtype is implemented instead (always with the possibility of recording alternative subtype information as a Note). In the case of localities, for example, the Biota fields [Locality]Latitude, [Locality]Longitude, [Locality]LatLongAccuracy, and [Locality]Elevation correspond to the POINT subtype in the ASC model. In other cases (e.g. the subtypes of COLLECTING-UNIT), the ASC subtypes are implemented using alternative strategies.

Some features of the ASC model not currently implemented in Biota are on the docket for implementation in a forthcoming version, for example, a full relational treatment for literature references. (Literature references can currently be accommodated as Notes.)

Biota's model includes three important features that are not included in the ASC model.

(1) Biota's [Determination History] table keeps track of all previous determinations of each [Specimen] record, while recording who changed the determination, when it was changed, and where it was changed (in the [Specimen] record, the [Species] record, or the [Genus] record). See Chapter 19 in the Biota manual.

(2) Biota's [Group] table implements a recursive relationship in the [Personnel] table, allowing designation of groups of collectors, determiners, borrowers, etc., based on permutations of [Personnel] records for individuals.

(3) Biota's Auxiliary Field system (based on a series of special tables) allows the user to create an unlimited number of custom fields for the [Species], [Specimen], [Collection], and [Locality] tables. See Chapter 15 of the Biota manual.

Annotated List of ASC Entities and Their Equivalents in Biota

The following list corresponds exactly to the alphabetical list of entities in the ASC document. ASC entities appear in uppercase (e.g. COLLECTING-EVENT). Biota entities (tables) appear in mixed letters within brackets (e.g. [Collection]). Biota field names are preceded by the table name (e.g. [Collection]CollectedBy). See the Biota Data Model for details of Biota tables, fields, and relationships.

ABSOLUTE-AGE-DETERMINATION: Not currently implemented in Biota.

AGENT (supertype): = [Personnel]. In Biota the [Personnel] table includes records for institutions (the ORGANIZATION subtype of AGENT in the ASC model), collecting "platforms" (the PLATFORM subtype of AGENT in the ASC model), and collections (the COLLECTION entity in the ASC model) as well as individuals. Through the [Group] table, Biota allows a one-to-many recursive relation between Group Personnel records and Individual Personnel records for members of each group.

CHRONOSTRATIC-AGE-DETERMINATION: Not currently implemented in Biota.

CITATION (supertype): Citations (references) are not currently implemented in Biota in normalized tables. Full citations can be entered for [Locality], [Collection], [Specimen], and [Species] records as text notes (any number per parent record) in the corresponding […Notes] tables ([Locality Notes], [Collection Notes], [Specimen Notes], or [SpeciesNotes]).

CLOCK-TIME (subtype): See TIME-EXPRESSION.

COLLECTING-EVENT: = [Collection].

COLLECTING-EVENT-ASSOCIATION: Association between [Collection] records for link [Specimen] records (e.g. parasites and hosts, plant parts and source specimen, DNA samples and donor specimen) are handled recursively through the [Collection] HostSpecimen attribute; see also COLLECTING-UNIT, below.

COLLECTING-EVENT-CITATION (subtype): See CITATION, above.

COLLECTING-METHOD: = [Collection]Method. The [Lists]Method field provides picklist support for [Collection]Method (see pp. 315-322 in the Biota manual).

COLLECTING-UNIT (supertype): = [Specimen].

SPECIMEN subtype. The SPECIMEN entity in the ASC model corresponds directly with Biota's [Specimen] table.

The LOT subtype. A LOT (a group of specimens of the same species from the same collecting event) is handled in Biota by setting the [Specimen] Abundance attribute in a single [Specimen] record to the lot count.

The UNSORTED-LOT subtype. An UNSORTED-LOT (a group of specimens of more than one species from the same collecting event) is treated in the same way as a LOT, but the [Specimen]SpeciesCode attribute for an unsorted lot record is set to the Temporary Taxon appropriate for the lot, e.g. "(Teleostes)" for an unsorted lot of teleost fishes; see Chapter 22, "Temporary Taxa for Approximate Determinations" in the Biota manual.

The SPECIMEN-COMPONENT subtype. In Biota, each SPECIMEN-COMPONENT receives its own [Specimen] record; the link to the donor (host) specimen is handled by recursively through the [Collection] HostSpecimen attribute. The rationale for this design is that the "collection" of the derived object from the donor specimen is a collecting event like any other, often with a different date, collector, method, etc. than the collecting event that produced the donor specimen.

The DERIVED-OBJECT subtype. In the ASC model, a DERIVED-OBJECT is an observation, image, or representation of a COLLECTING-UNIT. In Biota, text observations related to a [Specimen] record can be recorded in a [SpecimenNote] record linked (many-to-one) with the parent [Specimen] record. Images in Biota are currently supported (many-to-one) for [Species] records, but not for individual [Specimen] records.

COLLECTING-UNIT-ASSOCIATION: See COLLECTING-UNIT, above.

COLLECTING-UNIT-CITATION (subtype): See CITATION, above.

COLLECTION: = [Personnel]. In the ASC model, the COLLECTION entity records the names of research collections and the institutions they belong to. In Biota the [Personnel] table includes records for institutions and collections as well as individuals; see AGENT, above. (The [Collection] table in Biota serves a different purpose; see COLLECTING EVENT, above.)

COLLECTOR*: = [Collection]CollectedBy, linked many-to-one with [Personnel]; see AGENT, above.

DERIVED-OBJECT(subtype): See COLLECTING-UNIT, above.

DERIVED-OBJECT-TYPE: Not currently implemented in Biota.

DETERMINATION: = In Biota, each [Specimen] record specifies current determination data, consisting of the fields [Specimen]SpeciesCode (linked to [Species]SpeciesCode, [Specimen]DeterminedBy (linked to [Personnel]ShortName), and [Specimen]DateDetermined (plus [Specimen]DateDetFlag, to code for partial dates). Previous determinations for a [Specimen] record are saved in the [DetHistory] table (see Chapter 19 in the Biota manual).

DETERMINER*: = [Specimen]DeterminedBy; see DETERMINATION, above.

ESTUARINE-HABITAT-DESCRIPTION (subtype): See HABITAT-DESCRIPTION.

FRESHWATER-HABITAT-DESCRIPTION (subtype): See HABITAT-DESCRIPTION.

GEOLOGIC-AGE-DETERMINATION: Not currently implemented in Biota.

GEOLOGIC-TIME-UNIT: Not currently implemented in Biota.

GEOMETRIC-LOCALITY (subtype): see LOCALITY.

HABITAT-DESCRIPTION (supertype and all subtypes): Not explicitly implemented in Biota, but [CollectionNotes] would logically accommodate all such descriptions.

LINE (subtype): See LOCALITY.

LOCALITY (supertype, with a single subtype GEOMETRIC LOCALITY): = [Locality]. Of the three subtypes for GEOMETRIC-LOCALITY in the ASC model (POINT, LINE, and POLYGON), Biota implements only POINT, through four fields in the [Locality] table: [Locality]Latitude, [Locality]Longitude (both in decimal degrees, following Federal Information Processing Standard 70-1), [Locality]LatLongAccuracy, and [Locality]Elevation. Implicitly, circular polygons can be represented by specifying a value in [Locality]Accuracy. A second level of POINT locality data can be recorded in the [Collection]XCoordinate, [Collection]Y-Coordinate, and [Collection]XY-Accuracy attributes; see pp. 115-117 in the Biota manual.

LOCALITY-CITATION (subtype): See CITATION, above.

LOT (subtype): See COLLECTING-UNIT, above.

MARINE-HABITAT-DESCRIPTION (subtype): See HABITAT-DESCRIPTION.

NAMED-PLACE: The ASC model relies on this recursive place-name entity (linked to the LOCATION entity) to represent location hierarchies (e.g. town or city, county or parish, province or state, country). Biota accommodates named-place hierarchies in four re-namable fields of the [Locality] table: [Locality] Locality, [Locality]District, [Locality]State/Province, and [Locality]Country.

ORGANIZATION (subtype): See AGENT.

PALEO-COLLECTING-EVENT (subtype): Not currently implemented in Biota.

PERIOD (subtype): See TIME-EXPRESSION.

PERSON (subtype): See AGENT.

PLATFORM (subtype): See AGENT.

POINT (subtype): See LOCALITY.

POLYGON (subtype): See LOCALITY.

PREPARATION-ACTION: Biota records specimen preparation explicitly in two fields, [Specimen]PreparedBy (= PREPARATOR in the ASC model) and [Specimen]DatePrepared (with partial-date coding in the [Specimen]DatePrepFlag field) (= the PREPARATION-ACTION-DATE attribute of the ASC PREPARATION-ACTION entity). Additional information (corresponding to the ASC entities PREPARATION-METHOD and PREPARATION-STEP and their associated attributes) can be recorded in one or more SpecimenNote records in Biota. For many types of preparation, the [Specimen]Medium field provides additional, albeit implicit, information on specimen preparation.

PREPARATION-METHOD: See PREPARATION-ACTION, above.

PREPARATION-STEP: See PREPARATION-ACTION, above.

PREPARATOR*: See PREPARATION-ACTION, above.

REFERENCE-WORK: See CITATION, above.

RELEVANT-TIME (subtype): See TIME-EXPRESSION, below.

SPECIMEN (subtype): See COLLECTING-UNIT, above.

SPECIMEN-COMPONENT (subtype): See COLLECTING-UNIT, above.

SPECIMEN-COMPONENT-TYPE: Not explicitly implemented in Biota. As outlined above under COLLECTING-UNIT, components removed or separated from specimens (e.g. a skull from a mouse skin, seeds from a plant specimen, a DNA sample from a pickled lizard) receive their own [Specimen] records in Biota, linked to the [Specimen] record for the donor specimen through the "host-guest" system (Chapter 21 of the Biota manual). Information corresponding to SPECIMEN-COMPONENT-TYPE may appropriately be entered in the [Specimen]Medium field for the component's own [Specimen] record, or could be coded in the prefix for [Specimen]SpecimenCode (e.g. "SKULL00023456," "Seed00037264," and so on).

STORAGE-LOCATION: = [Specimen]Storage

STORAGE-MEDIUM: = [Specimen]Medium. If a specimen is transferred from one storage medium or storage regime to another (e.g. removed from alcohol storage, critical-point dried, and pinned), details may be recorded as a [SpecimenNote] in Biota.

STORAGE-REGIME: See STORAGE-MEDIUM, above.

TAXON-NAME: The ASC model uses an recursive, child-taxon/parent-taxon design for recording taxonomic names and relationships, and a complex many-to-many relationship between taxonomic names and taxonomic concepts (independent of names), to allow multiple, simultaneous classifications to coexist in the same database. Biota implements, instead, a simple, hierarchical design, based on 7 tables corresponding to the obligatory levels of the taxonomic hierarchy ([Species], [Genus], [Family], [Order], [Class], [Phylum], and [Kingdom]) to record taxonomic names and relationships. Biota treats intermediate taxonomic levels (e.g. Subfamily, Superorder) as attributes (fields) of the next lower entity (table) (e.g. [Genus]Subfamily, [Order]Superorder), with the exception of infraspecific taxa, which are recorded as [Species] table attributes (in Biota 1.1 and later). Fields for authors (authorities) are explicit for [Species]Author, [Species]SubspAuthor, and [Species]VarietyAuthor. User renamable fields in higher taxon tables can be designated as Author fields for higher taxa. Biota's design does not support multiple, simultaneous classifications.

In Biota, species-level synonymies are recorded through a recursive relationship in the [Species] table between [Species]SpeciesCode and [Species]ValidSpCode. (In the ASC model, synonymies are indicated in the TAXONOMIC-RELATIONSHIP-TYPE attribute of the TAXONOMIC-RELATIONSHIP table.)

TAXON-NAME-USE: See TAXON-NAME, above.

TAXON-NAME-USE-CITATION (subtype): See CITATION, above.

TAXONOMIC-CONCEPT: See TAXON-NAME, above.

TAXONOMIC-RELATIONSHIP: See TAXON-NAME, above.

TERRESTRIAL-HABITAT-DESCRIPTION (subtype): See HABITAT-DESCRIPTION.

TIME-EXPRESSION (supertype): In the ASC model, TIME-EXPRESSION handles temporal data for COLLECTING-EVENTS, with subtypes CLOCK-TIME (dates, times of day), PERIOD (non-date, non-clock time designations, e.g. season), and RELEVANT-TIME (non-date, non-clock, non periodic time designations, e.g. mating season). Biota records the timing of a collecting event using the fields [Collection]DateCollected and [Collection]DateCollEnd (the latter may be blank). Partial dates (year only, month-year) are accepted for both fields, flagged internally by the corresponding fields [Collection]DateCollFlag and [Collection]DateCollEndFlag. See pp. 111-114 in the Biota manual. Information corresponding to the intended use of the ASC PERIOD and RELEVANT-TIME subtypes may be entered as a [CollecionNote] record in Biota.

TIME-UNIT-BOUNDARY: Not currently implemented in Biota.

TRANSACTION: = [Loans]. In Biota, the [Loans] table may be used to record specimens (or lots) loaned to, borrowed from, donated to, or received as a gift or exchange from other individuals or institutions. See Chapter 17 of the Biota manual.

TRANSACTOR* = [Loans]Borrower, linked to [Personnel].

UNSORTED-LOT (subtype): See COLLECTING-UNIT, above.