Committee Meeting Fall 2008

To be held Nov 24, 2008 at 12:00PM noon.
Room 3027 of the CMMT
950 West 28th Avenue
Vancouver, B.C., V5Z 4H4

Comments from the Committee Meeting

  • Mark Wilkenson from iCapture
  • OMIM references - statistics?
  • Effectiveness of the scoring/Impact of the correction
    • Distribution of the scores - are many scores lost due to the p-value correction
  • method to better evaluate the strength of a gene-literature connections
    • number of genes tied to the paper (split votes)
    • term frequency, presence in title…
    • "strength" of a gene-paper connection vs. negative relationships
  • Remove uninformative terms like Human
    • weighting properties by predictive value?
  • Negatives can be guessed - presumed negative list?
    • Assume "mostly negative"
  • For comparison, show "random" Cum. Gains chart
    • need a better sense of what constitutes a "good" graph vs a "bad" graph
    • intuition to build a better scoring system
  • Sufficient work shown
  • Clustering genes
    • comparison of "known" gene-gene relationships (shown together before) vs. novel gene-gene relationships
    • new associations are more exciting
      • being able to evaluate "novelty" of a relationship - comparison to pre-existing evidence
  • drugs/chemicals - is PubMed annotation sufficient/good enough (life science bias?)
  • Show distribution of scores (and effect of correction)
  • "side projects" - fix deficiencies
  • start writing Intro - can become a (short) review paper)

Presentation Slides

Committee Meeting 2008 - Research Status Update.pdf (Updated Nov 21, 2003)

Research Progress

  • Data Sources: Entrez Gene, PubMed, Medical Subject Headings
  • Completed Preliminary Gene-Disease Prediction via Literature Model (Goal 1/3):
    • Direct Connections (Disease Terms associated to Literature directly referenced by gene)
    • Profile Comparison (Profile of terms associated with gene compared to profile of terms associated with disease)
    • Analysis of scoring metrics
    • Validation using OMIM and direct connections over time


  • MEDG 520 (Fall 2007)
  • All academic requisites complete


  • MSFHR SGS (starting Sept 2008): 2 years
  • Existing NSERC (Starting Sept 2007): 3 years


  • Warren Cheung and Ghassan Hamarneh. n-SIFT: n-dimensional Scale Invariant Feature Transform. Submitted to Transactions on Image Processing, under revision for resubmission.
  • Warren A Cheung, Francis BF Ouellette, Wyeth W. Wasserman. Mapping Biomedical Literature Annotations to Extract Gene-Disease Associations. Canadian Student Conference on Biomedical Computing (CSCBC) 2008.
  • Kemmer D, Podowski RM, Yusuf D, Brumm J, Cheung W, Wahlestedt C, Lenhard B, Wasserman WW. (2008) Gene Characterization Index: Assessing the Depth of Gene Annotation. PLoS ONE 3(1): e1440 doi:10.1371/journal.pone.0001440
  • Warren Cheung and Ghassan Hamarneh. Scale Invariant Feature Transform for n-Dimensional Images (n-SIFT). The Insight Journal, 2007 July - December, December 2007.
Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-Share Alike 2.5 License.