nodalpoint.org - Object-oriented Biology - Comments

The authors applied the model

Kevin — Mon, 13 Jun 2005 13:05:57 -0400

The authors applied the modeling language for software engineering to biological problems. The approach seems interesting, but invites many questions.

1) Is UML an appropriate way to describing biological processes?

2) As far as I know, there is no standard way of porting UML diagrams between different tools. How this approach can be useful if different UML tools are used?

3) The authors decribed the static characteristers with the class diagrams and dynamic features by the activity charts. How can both be implemented in the biological processes? Many tools translate the class diagrams into the static structures, but none of them can convert the activity charts into the logic inside methods.

4) In OOP, classes are templates and objects are instances. Is the same principle applied to the biological system? If so, what is a template and what is an object?

Kevin

biology<->code objects

Neil — Fri, 27 May 2005 23:56:30 -0400

This is very interesting stuff. I was giving a "what is computational biology" kind of talk to our school last year, and one definition that I came up with was "the process of converting biological objects to computational objects, performing operations and converting back again". Of course in my mind, I was thinking about "objects" in the programming sense (but you don't launch into a discussion of OOP principles with that kind of audience). I've always liked Bioperl because its designers seem to have a good understanding of this biological<->code object conversion and this article formalises it yet further.

No problem. Thanks Neil!

toniher — Fri, 27 May 2005 10:30:25 -0400

No problem. Thanks Neil!

edited post

Neil — Thu, 26 May 2005 23:23:07 -0400

hi toniher,

I edited your post as it was taking up most of the front page. The table can now be seen if you click "read more".

Neil

Object-oriented Biology

toniher — Thu, 26 May 2005 07:47:42 -0400

I have just read an interesting article which discusses about the application of object-orientation paradigm, used in popular programming languages such as Ruby, Java or C#, in dynamical and statical description of different gene products and their derived forms. Object attributes and properties, which could refer to proteins and their ulterior complexes, come from vocabulary terms created in Gene Ontology initiative.
This is a very suggestion approximation, since it can unite the different ontologies and organization principles of GO (Molecular Function, Biological Process and Cellular Component) in the object context.
This enable, for instance, to use class diagrams generated with UML, in order to describe and simulate biological processes, and particularly, the interaction of their components.
I reproduce from the paper the correspondence between object-orientation principles and the integration of the model with Gene Ontology.