Genes, information and sense: complexity and knowledge retrieval

Abstract

Information capacity of nucleotide sequences measures the unexpectedness of a continuation of a given string of nucleotides, thus having a sound relation to a variety of biological issues. A continuation is defined in a way maximizing the entropy of the ensemble of such continuations. The capacity is defined as a mutual entropy of real frequency dictionary of a sequence with respect to the one bearing the most expected continuations; it does not depend on the length of strings contained in a dictionary. Various genomes exhibit a multi-minima pattern of the dependence of information capacity on the string length, thus reflecting an order within a sequence. The strings with significant deviation of an expected frequency from the real one are the words of increased information value. Such words exhibit a non-random distribution alongside a sequence, thus making it possible to retrieve the correlation between a structure, and a function encoded within a sequence.

Chloroplast Genomes Exhibit Eight-Cluster Structuredness and Mirror Symmetry

“…For W (3,1) N = M , and it is not so in general case. A frequency dictionary W q of nucleotide sequences is claimed to be an entity bearing a lot of information on that latter [15][16][17][18][19][20]. A consistent and comprehensive study of frequency dictionaries answers the questions concerning the statistical and information properties of DNA sequences.…”

Lost Strings in Genomes: What Sense Do They Make?

“…Formula (5) looks like a Markov process expression, while it is not: it is derived with no hypothesis towards the Markov property of an origin sequence (see [1][2][3][4][5][6]for details). Thus, another idea to figure out a lost string is to distinguish "inevitably lost" strings from "unexpectedly lost" ones.…”

“…A frequency dictionary W q of nucleotide sequences is claimed to be an entity bearing a lot of information on that latter [1][2][3][4][5][6]. A consistent and comprehensive study of frequency dictionaries answers the questions concerning the statistical and information properties of DNA sequences.…”

Extracting DNA words based on the sequence features: non-uniform distribution and integrity

“…Some ab initio methods have been developed in the literature, such as k-mer [ 3 ], relative entropy [ 4 ], and information content [ 5 – 8 ]. In these methods, the frequency information of a word in a DNA sequence was used widely, but the position information was not paid enough attention.…”