r16 - 15 Oct 2006 - 18:01:40 - RobVogelbacherYou are here: TWiki >  Main Web  >  TWikiUsers > RobVogelbacher
Name: Rob Vogelbacher
Email: rjv_at_speakeasy_dot_net

Research Interests

Genetic Algorithms

  • A short paper on IBGGeneDesigner that I am submitting to the CTI Research Symposium as well as the Life Science Society's Computational Systems Bioinformatics conference (Aug 06.)
  • Here is a paper that I wrote on Genetic Algorithms and biological constructs: GA Paper
  • Currently (March 06) I am working on the development of a genetic algorithm for optimizing codon choice in synthetic gene design. I have finished the initial work on the GA and am currently incorporating functionality for the generation of oligonucleotides for Assembly PCR.
  • Planning on moving the IBGGeneDesigner over to ECJ in the near future. I'm particular excited about using the strengh-pareto optimization algorithm and exploring the use of island models on the cluster.
  • An abstract at ACM which describes a GA framework for the japanese biogrid.

Phylogenetics

  • Actively running MrBayes and PAUP* jobs for the folks at the Field Museum. Hoping to get an online job submission portal to further automate the process.
  • Dominic has some great ideas posted on the wiki already. Not sure if these will be pursued as he's changed his thesis topic and Dave is moving away from this type of research. Possibly worth investigating parallelization as described by Dominic. TODO: Transition phylogenetics work to Greg
  • Here's a powerpoint I put together for the group which discusses phylogenetics and MrBayes on z-cluster: Phylogenetics on Z-cluster

Proteomics

  • Previously, I assisted Dominic with making some data model changes to the MassSpec database. In the process, I came across a cool open source database tool called druid which helped me generate ERDs and system documentation for the database. In addition to the document generation (PDF, XHTML, XML) it also has plugins for Hibernate, Torque, MySql, postgres, and others. Already, I see myself becoing a big fan.
  • Created a prototype using JAXB for consumption of mzXML documents for the MS toolkit and Database project. I believe that Eric has incorporated this into his code for the MSIOP.
  • I've done some analysis of MS/MS data using Lutefisk and OMSSA. The area of de novo sequencing is particular interesting, as several authors have used graph-theoretic algorithms as the basis for their prediction models. I'd expect to see several competing de novo algorithms for the next breed of mass spectrometers provided the vendors don't get their first with their proprietary software.

Previosly Investigated

Bio-Grid Job Scheduling

  • I completed analysis of several possible job schedulers for the cluster in early July '05. The project participants and I agreed upon using Sun Grid Engine for job scheduling and resource management on Z. I was defining the topology (e.g. master host, submit host, etc.) for the installation when we decided to rebuild the cluster after a security breach. Instead of SGE, we are now using Torque which is bundled with the DCC distribution we used.

Lutefisk

  • Previously, I wrote a perl wrapper to execute Lutefisk with MS data in Mascot Generic Format (.MGF) and analyze the relevant results for automation of database searching using CIDentify. The script currently uses regexp matching to find "interesting" sequences in the Lutefisk output and then feeds it into CIDentify. The code is available in subversion here.

  • Here is the regexp to match Lutefisk output if anyone is interested:

$compiledPattern = "^(?:\\[\\d+.\\d+\\]|[ARNDCEQGHILKMFPSTWYVJOUXZ])+\\s+\\d\\s+\\d.\\d+";

OMSSA

  • I've run several OMSSA jobs both at the command line and via the web recently with limited success. Dave and Alex suggest that the proteins we've been looking for could in fact be novel which would explain our lack of results. TODO: Add execution instructions to project page.

ToDos

  • Create IBG keystore for automation of IBG Desktop jar out of subversion.
  • Evaluate use of automated building of IBG Desktop
  • Transition phylogenetics work to Greg Z.
  • Document IBG GeneDesigner on project page
  • Begin porting of GA to ECJ architecture

-- RobVogelbacher - 20 Mar 2006

Edit | Attach | Printable | Raw View | Backlinks: Web, All Webs | History: r16 < r15 < r14 < r13 < r12 | More topic actions
 
Illinois Bio-Grid
This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback