內容大鋼
遺傳可被視為是極長時間尺度上的基因組信息傳遞。然而，基因組會在像人類生命這樣短暫的時間尺度上產生偶發性錯誤（即突變），並可能導致顯著的影響。香農1948年發表的《通信的數學原理》開創了資訊理論這一新領域，證明了通過一個不可靠（即可能出現錯誤）的通道進行可靠的信息傳輸是可能的，但能可靠傳遞的信息有一個容量，超過了這個極限，可靠通信就不再可能了。近年來隨著Turbocodes等糾錯編碼的發明，在電子通信工程中以接近容量的方式可靠地傳輸信息已成為現實。我們日常生活的很多電子設備（如手機、CD、DVD、數字電視等）若沒有高效的糾錯碼是無法工作的。遺傳作為自然界在長久以前已解決的一個工程問題，也能用資訊理論的角度來研究。DNA的傳輸容量很容易以指數形式迅速衰減，這意味著必須使用糾錯碼來再生基因組，才能可靠地傳遞遺傳信息。假設這樣的糾錯編碼確實存在，就可以解釋生命世界的基本特徵，如分立物種及其等級分類，連續世代的必需性，甚至是向複雜生物進化的趨勢。本書有兩大目的，一是向遺傳學家展示資訊理論和糾錯編碼是遺傳信息傳遞的必要工具，還討論了使用它們的一些生物學結果，並提出了基因編碼的假設猜測。另一個目標是促使通信工程師對遺傳學和生物學產生興趣，以拓寬他們的視野，超越技術領域，向最傑出的工程師大自然學習。
目錄
Synthesis Lectures on Biomedical Engineering
Contents
Foreword
Ⅰ An Informal Overview
1 Introduction
1.1 Genetics and communication engineering
1.2 Seeing heredity as a communication process
1.2.1 Main and subsidiary hypotheses
1.2.2 A static view of the living world: species and taxonomy
1.2.3 A dynamic view of the living world: evolution
1.3 Regeneration versus replication
2 A Brief Overview of Molecular Genetics
2.1 DNA structure and replication
2.2 DNA directs the construction of a phenotype
2.3 From DNA to protein,and from a genome to a phenotype
2.4 Genomes are very long
3 An Overview of Information Theory
3.1 Introduction
3.2 Shannon's paradigm
3.3 Quantitative measurement of information
3.3.1 Single occurrence of events
3.3.2 Entropy of a source
3.3.3 Average m utual inform ation,capacity of a channel
3.4 Coding processes
3.4.1 Variants of Shannon's paradigm
3.4.2 Source coding
3.4.3 Channel coding
3.4.4 Normalizing the blocks of Shannon's paradigm
3.4.5 Fundamental theorems
3.5 A brief introduction to error-correcting codes
3.5.1 Redundant code,Hamming distance,and Hamming space
3.5.2 Reception in the presence of errors
3.6 Variant of Shannon's paradigm intended to genetics
3.7 Computing an upper bound of DNA capacity
3.8 Summary of the next chapters
Ⅱ Facts of Genetics and Information Theory
4 More on Molecular Genetics
4.1 Molecular memories: DNA and RNA
4.1.1 Unidimensional polymers as hereditary memories
4.1.2 Structure of double-strand DNA
4.1.3 RNA as another molecular memory
4.1.4 DNA as a long-lasting support of information
4.1.5 Error-correction coding as an implicit hypothesis
4.2 Place and function of DNA in the cell
4.2.1 Chromosomes and genomes
4.2.2 Principle of DNA replication
4.2.3 Genes instruct the synthesis of proteins
4.2.4 Amino-acids and polypeptidic chains
4.2.5 Synthesis of a polypeptidic chain
4.2.6 Proteins
4.3 Genome and phenotype
4.3.1 A genome instructs the development and maintenance of a phenotype
4.3.2 A phenotype hosts the genome from which it originates
4.4 DNA recombination and crossing over
5 More on Information Theory
5.1 Alphabet, sources, and entropy
5.1.1 Memoryless sources, Markovian sources, and their entropy
5.1.2 A fundamental property of stationary ergodic sources
5.2 About source coding
5.2.1 Source coding using a source extension
5.2.2 Kraft-McMillan inequality
5.2.3 Fundamental theorem of source coding
5.3 About channel coding
5.3.1 Fundamental theorem of channel coding
5.3.2 Coding for the binary sym metric channel
5.3.3 General case: Feinstein's lemma
5.4 Short introduction to algorithmic information theory
5.4.1 Principle of the algorithmic information theory
5.4.2 Algorithmic complexity and its relation to randomness and entropy
5.4.3 Sequences generated by random programs
5.5 Information and its relationship to semantics
5.6 Appendices
6 An Outline of Error-Correcting Codes
6.1 Introduction
6.2 Communicating a message through a channel
6.2.1 Defining a message
6.2.2 Describing a channel
6.3 Repetition as a means of error correction
6.3.1 Error patterns on repeated sym bols and their probability
6.3.2 Decision on a repeated symbol by majority voting
6.3.3 Soft decision on a repeated symbol
6.4 Encoding a full message
6.4.1 Introduction
6.4.2 A simple example
6.4.3 Decoding the code taken as example using the syndrome
6.4.4 Replication decoding of the code taken as example
6.4.5 Designing easily decodable codes: low-density parity check codes
6.4.6 Soft decoding of other block codes
6.5 Error-correcting codes within information theory
6.5.1 An outlook on the fundamental theorem of channel coding
6.5.2 A geometrical interpretation
6.5.3 Designing good error-correcting codes
6.6 Convolutional codes
6.6.1 Convolutional encoding
6.6.2 Systematic convolutional codes and their decoding
6.8 Historical outlook
6.9 Conclusion
Ⅲ Necessity of Genomic Error Correcting Codes and its Consequences
7 DNA is an Ephemeral Memory
7.1 Probability of symbol erasure or substitution
7.1.1 Symbol erasure probability
7.1.2 Symbol substitution probability
7.2 Capacity computations
7.2.1 Capacity computations, single-strand DNA
7.2.2 Capacity computations, double-strand DNA
7.3 Estimating the error frequency before correction
7.4 Paradoxically, a permanent memory is ephemeral
8 A Toy Living World
8.1 A simple model
8.2 Computing statistical quantities
8.3 The initial memory content is progressively forgotten
8.4 Introducing natural selection in the toy living world
8.5 E xample of a toy living world using a very sim ple code
8.6 Evolution in the toy living world;phyletic graphs
9 Subsidiary Hypothesis, Nested System
9.1 Description of a nested system
9.2 Rate and length of component codes
9.3 Distances in the nested system
9.4 Consequences of the subsidiary hypothesis
10 Soft Codes
10.1 Introducing codes defined by a set of constraints
10.2 Genomic error-correcting codes as 『soft codes』
10.2.1 Defining soft codes
10.2.2 Identifying the alphabets
10.2.3 Potential genomic soft codes
10.3 Biological soft codes form nested systems
10.4 Further comments about genomic soft codes
10.5 Is a eukaryotic gene a systematic codeword
11 Biological Reality Conform s to the Hypotheses
11.1 Genomes are very redundant
11.2 Living beings belong to discrete species
11.2.1 A genomic error-correcting code implies discrete species
11.2.2 Species can be ordered according to a hierarchical taxonomy
11.2.3 Taxonomy and phylogeny
11.3 Necessity of successive regenerations
11.3.1 Correcting ability of genomic codes
11.3.2 N ature must proceed with successive regenerations
11.3.3 Joint implementation of replication and regeneration
11.4 Saltationism in evolution
11.4.1 Regeneration errors result in evolutive jumps
11.4.2 Saltationism depends on the layer depth in the nested system
11.5 Trend of evolution towards complexity
11.5.1 Evolutive advantage of long genom
11.7 Relationship between genomes and phenotypes
11.7.1 Genome as specifying a phenotype
11.7.2 Neighborhood in genomic and phenotypic spaces
11.7.3 On genome comparisons expressed as percentages
12 Identification of Genomic Codes
12.1 Necessity of identifying genomic codes
12.1.1 An unusual approach
12.1.2 A necessary collaboration of engineers and biologists
12.2 Identifying error-correction means
12.2.1 Identifying an error-correcting code
12.2.2 Identifying component codes of the nested system
12.2.3 Identifying regeneration means
12.3 Genome distinction and conservation
12.4 Difficulties with sexual reproduction
13 Conclusion and Perspectives
Bibliography
Biography
Index
[