GeneFisher-P - User-Level Workflow Design: A Bioinformatics Perspective - page 94

4.3.1 Domain Model
The following describes the service and type taxonomies, service descriptions
and domain constraints that have been defined for the GeneFisher-P domain
model.
Taxonomi e s
Figure 4.8 shows the service taxonomy of the GeneFisher-P domain model.
The OWL class Thing is the root of the taxonomy, below which the EDAM
term Operation explicitly provides a group for concrete and abstract service
representations. The taxonomy comprises a number of service categories for
different operations concerned with the design, comparison, analysis and pro-
cessing of sequence data. As the figure shows, the concrete services (available
as SIBs) are then sorted into the taxonomy. The different file processing ser-
vices are only quite generally classified as Operation (upper part of the figure).
The actual GeneFisher services are then more specifically associated with the
DNA backtranslation , Sequence alignment conservation analysis , Multiple se-
quence alignment and PCR primer design classes (lower part of the figure).
The type taxonomy of the GeneFisher-P domain model is shown in Figure
4.9. It contains terms from both the Identifier and Data branches of the
EDAM ontology. While the former finally only contains the File name ,the
latter describes the URI data type more precisely, defines an explicit Sequence
alignment (multiple) term and comprises a number of categories for different
kinds of sequence data.
Services
Table 4.1 lists the services contained in the GeneFisher-P domain model (in
alphabetical order) along with their input and output type annotations in
terms of the EDAM ontology. The set of input types contains all mandatory
inputs (i.e., optional inputs are not considered), while the set of output types
contains all possible outputs. Note that the service interface definitions only
consider the data that is actually passed between the individual services, that
is, input parameters that are used for configuration purposes are not regarded
as service inputs. As also visible from the service taxonomy (Figure 4.8), the
GeneFisher-P domain model comprises the actual GeneFisher services for
sequence processing, alignment computation and PCR primer design, as well
as a number of services providing basic file processing functionality.
Constraints
Finally, a set of domain constraints is defined that covers general properties
of GeneFisher-P workflows, such as that the GeneFisher service that actually
designs the PCR primers is always included, that the individual analysis steps
are not repeated unnecessarily, or that steps are executed in a particular order: