Next Meeting: Mo, 29.06., 17h, UV10

HMM Based Evolution

The idea is to use Transducers, which act like pairHMM for simulating evolutionary events. Those Transducers should be based on grammars, so there also a need for a specific grammar and its parser.



Useful Web Ressources

First Ideas

  • One Superannotation for all possible annotations
  • User can create new Anootations using the Superannotation
  • All evolutionary steps will be built from a set of parametririzable basic operations:
    • Insertion
    • Deletion
    • Duplication
    • Double-Cut-Join
    • Move (Transposon)
    • Substitution
  Transducer is sequence based 
  * Need of neutral state (no event) 
  * Step by step (base by Base) 
  Hierarchical annonations, tree like -- (Chromosome (Gen (Intron, Exon)), Transposon) 

Further Bla

  • Mutationprobabilities (edges in phylotree (inverse probabilities) must be >= 1) will affect only outgoing edges from Match state
  • Probabilities will be stored in matrices for each annotation and mutation
  • Startingposition for Transposon/Satelites will be set automatically

Use Cases

  • Automaton Modell definieren √
  • Automaton Schnittstellen definieren & implementieren√
  • Automaton fuer Character-Basierte Evolution erstellen √/X
  • Automaton Controller erstellen √/X
  • Join Automata √
  • Import von konkreten Automata erstellen X
  • Schnittstelle zum User Interface herstellen X
  • Konkrete Automata fuer verschiedene Annotationen erstellen X

knapp 50% √

Final Report

Download here .