A Tour of Molecular Information Theory:
Suggested reading

Thomas D. Schneider

This page develops the application of information theory to biology by topic in roughly historical order. You can read every paper in the order given. Most papers, including not ones listed here, can be found on the page Molecular Information Theory Papers a open book Molecular Information Theory Papers. If you need to look up terms, you can use the glossary.

All papers will open in a separate window. The glossary also has its own window.

Shannon uncertainty equation: H equals the negative
sum from i equals 1 to M of P sub i log 2 p sub i bits per
symbol.

Basics

glossary A list of terms used throughout this website.
primer Information Theory Primer With an Appendix on Logarithms.
Basic information theory in a nutshell

Average Information in Nucleic Acid Binding Sites

schneider1986 The information in a binding site is just enough for it to be found in the genome.
logopaper Sequence logos, a graphical display of binding sites.
ev Evolution of binding sites, be sure to run the Evj program!
correlogo 3D sequence logos

Individual Information in Nucleic Acid Binding Sites

ri Theory of individual information
walker Graphical display of individual information
zen Consensus Sequence Zen. What's wrong with consensus sequences?

Discoveries

zheng.storz2001 Sequence walker predicted OxyR binding site confirmed by DnaseI footprint
Flipping Bases out of DNA
- baseflip Evidence for base flipping from sequence logos, the first step of DNA replication
- repan3 Testing the base flip hypothesis experimentally.
Fis sites and Molecular Computers
- fisinfo Analysis of Fis binding sites
- fff fisflipflop: Discovery that Fis binding sites overlap to form a flipflop.
- moco Molecular Computing Elements: Gates and Flip-Flops. Pushing the discovery of overlapping Fis binding sites to create a molecular computer.

Fis Flip Flop - two adjacent Fis protein binding sites
compete for Fis protein binding

Molecular computer circuit diagram for a NOR gate made
from one repressor protein blocking two different activator
proteins.

Flexible binding site models
- flexrbs E. coli ribosome binding sites can be described by a flexible model.
- flexprom Applying the flexible model to E. coli sigma⁷⁰ promoters

T7 promoters and islands
- t7like Discovering promoters of T7-like bacteriophages
- t7island Discovering bacterial genetic islands that use T7 promoters

Thumbnail for Chen Schneider 2005 paper, phylogenetic
trees for T7-like sequence logos and T7-like RNA
polymerases.

thumbnail for Chen Schneider paper on T7 islands 2006
figure 8 Alignments of T7 islands.

Thumbnail for Chen Schneider paper on T7 islands 2006
figure 9. Phylogenetic trees showing recombination between
T7 islands.

Clusters
- fur Discovery that Fur sites come in clusters

Analysis of Diseases

colonsplice Using Information Content and Base Frequencies to Distinguish Mutations from Genetic Polymorphisms in Splice Junction Recognition Sites
rfs Analysis of splice junction mutations that cause diseases
abcr Medical Applications of Sequence Walkers
splice server citations (PDF) Papers that cite the splice analysis server for understanding genetic diseases

Shannon channel capacity equation: C equals W log base
2 of P over N plus 1 bits per second.

Advanced Theory

ccmm Channel capacity of molecular machines. Up to this point sequence logos are about patterns measured in bits. This paper links bits to energy dissipated during binding.
edmm The Second law of Thermodynamics is the relationship between energy and information.
shannonbiologist Why information theory is a theory about biology.
fisbc More about Fis binding sites and the first experimental evidence showing that information theory is a theory about biology because it correctly predicts a breakpoint in the DNA binding versus information curve.
emmgeo A fundamentally important question in molecular biology is: What is the the relationship between energy and information?
To set a coin to one state, it must dissipate energy. The amount of energy can be larger or smaller, but it has a minimum for one bit. So we can convert energy to the maximum bits that could be gained for that energy dissipation. This paper compares the actual information to the energy dissipation, measured in bits, to get an efficiency. Unlike the famous Carnot efficiency, which depends on a temperature difference, this is an isothermal efficiency. The paper shows that the efficiencies of many molecules are near 70%. It then shows why using a high dimensional space representation of the molecular states.

Philosophy

Pitfalls in Information Theory and Molecular Information Theory. Fall into any one of these traps and you can't get the rest of the theory! (This is not a paper but might be some day.)
ridebate Why one should not use "relative entropy". See also Using `Relative entropy'.
Recommendations for Making Sequence Logos. Get more out of your logos.

Reviews

hawaii 1993 Basic information theory
nano2 1994 Review of ccmm and edmm.
lessons2003 2003 Review of lessons learned
schneider2006 2006 Review: sequence logos, sequence walkers
brmit 2010 Review through to efficiency

color bar Small icon for Theory of Molecular Machines: physics,
chemistry, biology, molecular biology, evolutionary theory,
genetic engineering, sequence logos, information theory,
electrical engineering, thermodynamics, statistical
mechanics, hypersphere packing, gumball machines, Maxwell's
Daemon, limits of computers

Schneider Lab
origin: 2007 Dec 11
updated: 2011 May 26

color bar

A Tour of Molecular Information Theory: Suggested reading