Slowpoke Comics.


Here is a workshop I am co-organizing with Prof. George Karypis

1st International Workshop on Data Mining in Drug Discovery (RxDM ‘09)    held in conjunction with SIAM Data Mining Conference (SDM 2009)


 Dates:  April 30- May 2, 2009 Sparks, NV, USA

 Drug discovery is an expensive and challenging process. In silico methods have played a key role in aiding the discovery process, in pharmaceutical companies and government institutes alike. Government initiatives (Molecular Library Initiative and Chemical Genetics Initiative) have lead to an increase in the amount of available public data in the form of chemical compounds, and experiments analyzing the interaction of compounds with proteins/genes i.e., biological spaces. In conjunction, the advances in sequencing technologies have lead to similar increases in the amount of sequence data in the form of DNA, genes, and proteins. Biological experiments related to gene expression profiling, and structural genomics initiatives have also added another dimension of data. There is a need for mapping the different spaces “biological” with “chemical” to understand the function of complex biological systems, with implications in drug discovery.

Suggested Topics (but not limited to the following) include:

   1. Structure-Activity Relationship Models.

   2. Chemical Descriptor Spaces.

   3. Predictive Toxicology.

   4. Gene, Disease, Drug Connections.

   5. Chemical genetics and  chemogenomics

   6. Structural Bioinformatics

   7. Systems Biology and Drugs

   8. Network Pharmacology

More information:

Contact Workshop Chairs at rangwala AT CS DOT GMU DOT EDU for questions. 

Chairs: Huzefa Rangwala (George Mason University) and George Karypis (University of Minnesota)

I was made aware of using “jigsaws” as an effective active learning strategy at a Preparing Future Faculty class I took at the University of Minnesota (my alma mater). The jigsaw is a collaborative learning activity where students learn a topic of interest by splitting the task. The topic or class module is split amongst students, so that a set of students are responsible for a particular sub-topic. The students are expected to read, and study the sub-topic at sufficient depth so as to call themselves as experts of the particular topic. During class students first convene in expert groups, discuss the sub-topic amongst themselves. After strengthening their ideas about the particular topic students are asked to gather in mixed groups such that every group has at least one experts from all the sub-topics. Collectively each mixed group talk about the entire topic, discussing and learning from others.

I have tested this activity in two classes that I have been instructor for, and have found sufficient enthusiasm amongst the student population to pursue it next time as well. I introduced students to five multiple sequence alignment algorithms  in a bioinformatics class taught at University of Minnesota, and five clustering algorithms in a data mining class taught at George Mason University. The main advantage foreseen was that students gained information about five different methods by reading only one of the research papers. Here is an example of this activity for the clustering algorithms.

The author of this post, Huzefa Rangwala is an assistant professor Computer Science Department, George Mason University.