Minutes 07/03/06

Project Rearrangements in genomes with unequal content
Date 07/03/06
Version 1.0
Purpose of Meeting Finalize experiments
Supervisor present Leong Hon Wai, Guillaume Bourque

Current Task/Sub-Task

  • Complete draft of report
  • Finish running experiments

Reported On

  • Experimental results for simulated data
  • Ideas on how to avoid removing genes in the MGR-GT algorithm

Discussed

Additional experiments

Run experiments to investigate the following:

  • scalability of the method, fixed k and %del, vary number of genomes
  • on a real dataset of 4 genomes from Dr Bourque
  • compare against GoTree and GRAPPA
  • show the false postitive for the dataset with outgroups

Presentation of results

Prof Leong commented that it would be better to use a common scale for all plots for a particular experiment.

Propagating gene content

At the moment, when merging two genomes G1 and G2, the gene content of the merged genome is taken to be the intersection of the content of G1 and G2. It may be possible retain more genes in the merged genome by considering a difference scheme of merging.

For example if G1 = [2,1,5,6] and G2 = [2,5,3,6] the merged genome Gm can be [2,1,5,3,6] since we can obtain information about the position of gene 1 and 3 from G1 and G2 respectively. However if G1 = [2,1,5,6] and G2 = [2,3,5,6] then it is not clear whether Gm should be [2,1,3,5,6] or [2,3,1,5,6].

Alternatively in the MGR-GT algorithm we are always solving the median problem for G1, G2 and G3. Suppose G1 = [2,1,5,6], G2 = [2,3,5,6] and G3 = [1,5,3,6,2] then we can let Gm = [2,1,3,5,6] where we include 1 and 3 in the median because it is present in 2 out of 3 genomes and the order of 1 and 3 in Gm can be determined by looking at their order in G3.

Things to report on in next meeting

  • Results of additional experiments
 
mgr/mtg_0019.txt · Last modified: 2007/12/29 09:37 (external edit)
 
Recent changes RSS feed Creative Commons License Donate Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki