About distance metrics for biological sequences.
Determining trees from data
Here is a new complete version of the notes (28 Feb 08).
Here is a revision of Chapters 7 and 8
Here is the latest revision of the first part of the notes, and here are the references.
This course will explore the digital nature of biology at the molecular scale. We will focus on the role of the hydrophobic effect in protein/ligand associations.
Protein interactions are discrete in nature even though hydrophobic effects are non-specific in general. There is a useful analogy with the duality between the analog and digital nature of computer chips. We refer to this study as the Digital Biology Project. We pursue basic biophysical issues but we also apply our ideas to biomedical problems, e.g., to contribute to the understanding of antibody binding and to drug design.
The course will primarily utilize data-mining as a tool both to understand basic biophysics and to explain protein-ligand associations.
The course will explore the connections among the following well known but seemingly contradictory facts:
The course will describe the use of datamining to understand the biophysics of proteins, and we will review the derivation of basic models for dielectrics. We will also discuss some changes at the quantum level caused by the large-scale hydrophobic and electronic environment. Implications for protein folding models will be discussed.
No particular background will be assumed. All prerequisites will be provided in the class. We will explore issues in computer science, applied mathematics, physical chemistry and biomedical applications. It is hoped that people from different disciplines will participate.
Notes for the course will be distributed and students will read some primary journal articles as well.
Here is the pdf file for the first two lectures.
Some Journals for the course.