Probability Review

conditional independence
posterior probability
prior probability

JavaBayes

JavaBayes is coded in Java. The JavaBayes site includes download information, some examples (dog, alarm networks), and a brief introduction to JavaBayes. The feature-set for this software is not very large.

It allows creation of networks using a GUI environment. The GUI environment consists of a network editor with various modes: create, move, delete, query, observe, edit variable, edit function and edit network. A console window logs the actions that take place and also acts as an output window. After setting various observations and toggling explanatory nodes, the posterior marginal, posterior expectation, estimation of explanatory variables, finding the complete explanation can be calculated for a particular node. The software supports various formats: BIF, XML and BUGS.

Nir Friedman

He does a lot of research using Bayesian networks. I don't understand the mathematics of the underlying probability, but I will keep digging.

I'm having lots of difficult understanding the probability terms. I will need to learn more about Bayes and Bayesian networks. It seems Friedman has the most work in this area so far.

Bayesian Network Software

A table of available software for Bayesian networks can be found at this site: Software Packages for Graphical Models/Bayesian Networks (2005). Here is Google's list of belief network software.

This looks like a great site for Bayesian Artificial Intelligence with notes, talks and course materials.

Michael Zhang

Zhang is a part of this research project: Genomics of Transcriptional Regulation. The site seems a bit old, but it has some software related to gene regulatory networks.

Here are some of the papers from his publications list:

1. Zhang MQ (2005) Inferring Gene Regulatory Networks - Chapter 21 in: Bioinformatics - From Genomes to Therapies (T. Lengauer, ed.) Wiley-VCH – in press.
2. Banerjee N, Zhang MQ (2002) Functional Genomics as Applied to Mapping Transcription Regulatory Networks
   
   • 5 pages
   • survey of the recent happenings for gene transcription regulatory experiments
   • experiments on the yeast cell
   • concentrates on cis-regulatory elements
   • discusses clustering approaches to classify certain genes
     
   • Bayesian networks are discussed in reference to Friedman
3. Banerjee N and Zhang MQ (2005) Transcription Regulatory Networks in Yeast Cell Cycle - Chapter in: Microarray Expression Profiling and Transcriptional Regulatory Networks (F Shannon and S Rao, eds.) In Press.
   
   • 12 pages
   • this seems like a more mature version of the paper above
     

Bayesian Networks

My research topic is Bayesian networks and their application to gene regulatory networks. Some data-mining may also be used. Here are some introductory links to Bayesian networks: Wikipedia - Bayesian networks, An Introduction to Bayesian Networks and their Contemporary Application and A Brief Introduction to Graphical Models and Bayesian Networks.

Michael Zhang's Computational Biology and Bioinformatics Lab
Hao Li's Lab
Wei Wang's Home Page
Mark Gerstein's Lab
Nir Friedman's Home Page

First Class

The first class for Dr. Wang's class was very different. The class consists of about 5 students from varying backgrounds. Some are from a mathematics background, some from chemistry, and some from physics. The professor wants to potentially cover protein folding, protein dynamics, biomolecular interactions and recognition, electron and proton transfer, motors, single molecules and single cells, cellular networks, development and differentiation, brains and neural systems, and evolution.

The grade for the course is soley determined by performance on a term project. Based on a brief biography sheet we filled out in the first class, Dr. Wang assigns us a project to work on throughout the semester. This seems to be a research-heavy course.

Biological Physics and Biophysical Chemistry

Looking through the job postings at BioSpaceJobs (link from ColorBasePair), many of the jobs require some BioTechnology, BioMedical, or BioChemistry background. Dr. Jin Wang is teaching a course at Stony Brook University called Biological Physics and Biophysical Chemistry: Theoretical Perspectives. Information about the course is available on the Computational Biology web page in the Computer Science department.

A course of interest being taught in Spring 2006 is Phys. 680.2/Chem. 683/AMS. 680, "Biological Physics and Biophysical Chemistry: Theoretical Perspectives", taught by Jin Wang. This course will survey a selected number of topics in biological physics and biophysical chemistry. The emphasis is on the understanding of physical organization principles and fundamental mechanisms involved in the biological processes. The potential topics include: Protein Folding, Protein Dynamics, Biomolecular Interactions and Recognition, Electron and Proton Transfer, Motors, Membranes, Single Molecules and Single Cells, Cellular Networks, Development and Differentiation, Brains and Neural Systems, Evolution. There will be no homework or exam. The grades will be based on the performance of the term projects.

Resource Hubs

BioPlanet and ColorBasePair seem to be good resources for BioInformatics. BioPlanet publishes a list of BioInformatics-related companies. This company list could be a good start. ColorBasePair has links to various job search sites specifically for BioInformatics jobs. Much of the resources defining BioInformatics, degrees, and FAQs are duplicated on both sites.