Food & Resource Economics
University of Delaware
|Topic:||Assessing Gene Expression measurements|
DNA microarrays are capable of providing genome-wide patterns of gene expressions across many different conditions over time. The analysis of these patterns requires detecting whether observed differences in expressions are significant or not. Normalization is the term used to describe the process of removing noise and variability of the data. Current methods are unsatisfactory due to lack of systematic framework that can accommodate such noise, variability, low replication of microarray data and missing value estimation.
We developed a parametric likelihood framework for microarray data analysis. We included missing data in the model, using appropriate techniques. We used different estimation techniques in order to validate the parametric likelihood model and the parameters of the model in the actual data set. Furthermore we used bootstrapping to estimate the variances of the estimated parameters.
Based on the parametric likelihood model we developed clusters of gene expression levels using principal component analysis, k-mean clustering and variance clustering techniques.
|Thursday, November 15, 2001 |
|Time:||Social hour -- 6:00 pm|
Dinner -- 6:30 pm
Speaker -- 7:15 pm
|Place:||Newark Holiday Inn|
Oliver's Restaurant map
|Menu:||Appetizer: Fruit Cup|
Entree: Chicken Oscar
Dessert: Ice Cream
If you desire a vegetarian meal, please so indicate in your RSVP.
|Cost:||$25 for members, $10 for students|
Reservations to George Chao by Friday, October 12, 2001.
302-992-5135, George Chao.
Please note: Be sure to RSVP by this date, to assure a meal is reserved for you.
Return to Delaware Chapter ASA home page.