Data Tracking Systems - Biomedical Informatics in Translational Research - page 130

but should be professional software developers. A couple of biomedical
informatics or bioinformatics scientists on the team can bridge the gaps
between the end users and the developers. They typically can understand and
translate the lab needs into IT language, but without systematic IT training
they are not in a position to develop a solid application system.
To develop a data tracking system with a partner company—typically an
existing LIMS company—a couple of business models can be used. One is to
follow a vendor-customer model, in which case the vendor (LIMS company)
provides professional services to the customer to specifically develop the
modules for the customer and maintains and supports the custom modules
thereafter. In this model, the costs of development can be high because there
is no future possible profit coming out of the development for either the
vendor company or the customer.
Another relationship is a partnership, in which case the developer
company and the customer organization form a partnership, and the
developer company uses the customer as an example case in the development
of the data tracking modules. These modules are developed in such a way
that they can be adapted to serve other customers. In this way, both partners
can benefit from the resale of the modules in the future and therefore
alleviate the cost of the development in the long run.
To develop a data tracking system with a partner company, three types of
team members are still needed, in which case the end users are from the
customer, the developers are from the developer company, and the
biomedical informatics or bioinformatics scientists can be from either
organization. In addition, a project management team should be formed,
with one project manager from either party. Furthermore, because two
organizations are involved, communications between the two parties will be
very important. It is also critical to understand the common needs and the
distinct interests of the two parties so that both parties will benefit from the
partnership and end up with a win-win outcome for the project.
2. System requirement collection. The system requirements described in Section
7.3 are a good starting point. For development purposes, the collection
requirements should be highly detailed and very specific to the operational
needs, which will be an essential component of the use case specification for
development. For example, for an experimental platform, the whole
experimental workflow should be described, and for each step, the operation
that is to be tracked and the data types that need to be tracked should be
specified. How the data from an experimental instrument are stored should
also be specified: simply uploading the original output data file, or parsing
the output file to store selected data elements, or simply storing the file
location or using a hyperlink. This requirement collection should be done for
all data platforms and experimental platforms to be covered. The collection
requirement needs detailed input from future end users, and should be
documented by the intended developers to make sure that all of the needed
information is collected.
The requirements also include the projected number of users of each data
platform, the number of concurrent users when the system is in full