biogram (version 1.0)

biogram-package: biogram - analysis of n-grams from biological sequences

Description

biogram package specialises in analysis of n-grams from biological sequences as nucleic acids or proteins.

n-grams

n-grams (k-tuples) are sequences of n items derived from longer sequences. They may be continuous or not. From example, from the sequence of nucleotides AATA one can extract following continuous 2-grams (bigrams): AA, AT and TA. Moreover, there are possible two bigrams with a single space between elements A_T and A_A and one bigram with double extra space between elements: A__A. In the biogram package, count_ngrams function is responsible for extracting n-grams. The d argument is responsible for specifying the distance (extra spaces) between the elements of the n-gram.

Another important parameter is position. Instead of counting just positive n-grams, one may count how many positive n-grams occur at given position. For example in the sequence of nucleotides AATA at position 1 is only one 2-gram AA (in the biogram notation 1_AA). That means, that all other possible bigrams 1_AC, 1_AG, 1_AT, 1_CC and many more are not present. Such data can be extracted using count_ngrams function with pos parameter.

n-gram data dimensionality

n-grams obviously suffer from the curse of dimensionality. For example, the peptide of length 6 has $20^{n}$ n-grams and $6 \times 20^{n}$ positioned n-grams. Such enormous data is very hard to manage in R. Furthermore, it is hard to perform analysis in that big feature space.

biogram package deals with both of mentioned problems. It uses innate properties of n-gram data (usually a very sparse matrices) to store them using functionalities provided by the slam package. To ease the choice of significant features, biogram provides user with the QuiPT, very fast permutation test for binary data (see test_features).

Another way of reducing dimensionality of the problem is the aggregation elements into bigger groups. For example, all positively-charged amino acids may be aggregated into one group. This action can be performed using the degenerate function.

Examples

#use data set from package
data(human_cleave)
#first nine columns represent subsequent nine amino acids from cleavage sites
#degenerate the sequence to reduce the dimensionality of the problem
deg_seqs <- degenerate(human_cleave[, 1L:9],
                      list(`1` = c(1, 6, 8, 10, 11, 18),
                           `2` = c(2, 13, 14, 16, 17),
                           `3` = c(5, 19, 20),
                           `4` = c(7, 9, 12, 15),
                           '5' = c(3, 4)))
#extract bigrams
bigrams <- count_ngrams(deg_seqs, 3, 1L:4, pos = TRUE)