1. IsoGene Projects: http://www.ibiostat.be/software/IsoGeneGUI/index.html
- The IsoGene package: https://r-forge.r-project.org/projects/isogene/
- IsoGene paper: http://journal.r-project.org/archive/2010-1/RJournal_2010-1_Pramana~et~al.pdf
- IsoGeneGUI package http://www.bioconductor.org/packages/release/bioc/html/IsoGeneGUI.html, chapter 17 of book Modeling Dose-Response Microarray Data in Early Drug Development Experiments Using R: http://link.springer.com/chapter/10.1007/978-3-642-24007-2_18
2. BiclusteringGUI project: http://www.ibiostat.be/software/BiclustGUI/index.html ,
3. Model-based and Model Averaging dose-response microarrays, More detail please see chapter 13 of book : Modeling Dose-Response Microarray Data in Early Drug Development Experiments Using R, springer
4. Multiple Contrast Tests for testing for trend in dose-response microarrays
5. Hierarchical Bayesian, Gibbs Variable Selection.
My research focus is two-fold. I am involved in methodological and applied work involving modeling of microarray data and preparing user friendly software packages in R. The first research topic I am currently involved is the dose-response modeling in Microarray Experiments. This study focuses on testing whether there is any evidence of the drug effect, investigating the nature of the dose-response relationship and then estimating the target doses (i.e., ED50 and Minimum Effective Dose) within microarray settings. To incorporate for model uncertainty in the set of possible dose-response models, model averaging is carried out to estimate the ED50 and MED.
The other area of interest that I am currently involved in is the Multiple Contrast Tests (MCTs). This research is aimed to find difference in means of gene expression between higher doses and the control. We apply multiple contrasts test to several test statistics, such as Williams, Marcus’ the M and the modified M tests. The MCTs SAM (Significance Analysis of Microarray) procedure is also proposed and investigated. These proposed statistics are found to have higher power to detect a monotonic trend.
I am also involved in a hierarchical Bayesian modeling approach for dose-response microarray experiments. In this research, the Bayesian approach seeks evidence in the data under both the null and alternative hypotheses under the assumption of monotonic. This is made possible by fitting all possible competing models under both the null and alternative hypotheses and selects the most suitable model for each gene using Gibbs Variable selection method.
In addition to methodological research, I am also involved in preparing and maintaining several R packages. The first package is an R package for dose-response in microarray: IsoGene which is available in CRAN. In addition to the IsoGene, I am developing the IsoGene Graphical User Interface (IsoGeneGUI) R package which provides the users with a flexible and user-friendly application for dose-response microarray analysis. The package is now available in the Bioconductor:
Besides research in dose-response, I am also involved in biclustering for microarrays research. In this research I am developing user friendly software for several Biclustering methods called BiclustGUI. The package is built as an extension for R-Commander (Rcmdr Plug-in). It is still in early development and available at https://r-forge.rproject. org/projects/biclustgui/.
In the beginning of my PhD, I am also involved in the bioinformatics aspect of mathematical and Statistical model for Infectious Disease project. The study is to infer “who infected whom“ in outbreak of a particular Infectious disease using the genetic distance of the virus. Likelihood-based estimation of Reproduction number (R) is carried out to obtain the probability that one person has been infected by another person. Based on this result, then we can construct the transmission network.
No comments yet.