Feb 28th UBM meeting at 12:00 invited talk:

 

Finite splicing systems, logically and algebraically

Patrick O'Neill

Towson University, Department of Mathematics

 

In the past fifteen years, many different models of DNA computation have been proposed, and universality results for many of these constructions have been immediate.  Not all models, however, are so easily characterized.  In particular we survey "finite splicing systems", models which compute through the action of restriction and ligation enzymes on initial strands of DNA.  Although finite splicing systems are among the oldest and most biologically relevant models of DNA computation, they have not been shown to conform to any known  computational class.  After comparing their expressive power to that of alternative models, we review what is currently known about finite splicing systems by presenting partial characterizations of splicing languages in terms of several canonical sub-regular classes, and describing algebraic and model-theoretic lenses for viewing the problem.

 

Reference literature: