Related Experiment Videos
Entropy and predictability of information carriers
1Institut of Physics, Humbolt-University, Berlin, Germany. werner@summa.physik.hu-berlin.de
Bio Systems
|July 2, 1998
Summary
This study uses entropy concepts to analyze information-carrying strings like DNA and proteins. Findings reveal long-range correlations in these structures, enhancing our understanding of information organization.
Area of Science:
- Information theory
- Bioinformatics
- Computational linguistics
Background:
- Linear strings like DNA, proteins, and text encode vast amounts of information.
- Understanding the underlying structure and organizational principles of these informational strings is crucial.
- Entropy concepts offer a powerful framework for quantifying information and complexity.
Purpose of the Study:
- To investigate the structure of information-carrying linear strings using entropy concepts.
- To explore the relationship between order, predictability, and information content.
- To analyze the applicability of these methods to diverse information carriers.
Main Methods:
- Introduction and generalization of conditional entropy and transinformation.
- Analysis of mutual information functions for biological sequences (virus DNA, proteins).
- Formal comparison with textual and musical string structures.
Main Results:
- Demonstration of entropy concepts' capability to describe information carriers.
- Identification of long-range correlations in various informational strings.
- Quantification of the link between order and predictability in these sequences.
Conclusions:
- Entropy-based analysis provides a unified approach to understanding diverse information carriers.
- Long-range correlations are a common feature in biological, textual, and musical information.
- The framework offers insights into the fundamental principles governing information organization.