GeneFisher-P - User-Level Workflow Design: A Bioinformatics Perspective - page 90

the input consists of multiple nucleic or amino acid sequences. Depending on
the type of input, services for sequence alignment computation, consensus cal-
culation, backtranslation, and finally for the primer design itself are invoked
by the process. Figure 4.4 summarizes how the different kinds of inputs are
processed before the actual primer calculation based on a single nucleotide
sequence can take place: Multiple input sequences have to be aligned and
transformed into a (single) consensus sequence, protein sequences have to be
translated (back) into an equivalent nucleotide sequence.

sequence
alignment

consensus
calculation

multiple protein

backtranslation

sequence
alignment

consensus
calculation

multiple DNA

primer
output

primer
design

sequence
input

single protein

backtranslation

single DNA

Fig. 4.4 Abstract GeneFisher workflow (following [124])

GeneFisher is a completely predefined application. Its internal workflow
is hidden from the end user, who can only interact with it through the web
GUI. The limitations of this interface become obvious in particular when it
comes to the full automation of the primer design process (batch processing)
or the exchange of single components. The objective behind GeneFisher-P
(the -P denotes the process-based realization of the application) was to put
the processes in the foreground and expose the internal workflows and the
underlying services and components to the end user, who becomes then able
to intervene and change or integrate them with others according to his specific
needs (examples are given in Section 4.2.4).

4.2.1 GeneFisher Services

The tools that are used by the GeneFisher web application and likewise for its
GeneFisher-P workflow realization, such as the alignment programs ClustalW
and DCA, are all hosted at BiBiServ. Some of them provide web service
interfaces, others are legacy programs that have been integrated using the
jETI technology (cf. Section 2.1.2). These tools originate from the first days of
the GeneFisher project, have been written in C and compiled for a particular
CPU type, and are thus dependent on specific machines, which means that
they can cannot be made available via the internet directly. The following
services are central within GeneFisher-P:

•

ClustalW (cf. Section 3.1.3), which is available as web service at BiBiServ.