XString-class: BString objects
In anandhupresannan/biostrings: Efficient manipulation of biological strings

Description Details Constructor-like functions and generics Accessor methods Coercion Subsetting Equality Author(s) See Also Examples

The BString class is a general container for storing a big string (a long sequence of characters) and for making its manipulation easy and efficient.

The DNAString, RNAString and AAString classes are similar containers but with the more biology-oriented purpose of storing a DNA sequence (DNAString), an RNA sequence (RNAString), or a sequence of amino acids (AAString).

All those containers derive directly (and with no additional slots) from the XString virtual class.

The 2 main differences between an XString object and a standard character vector are: (1) the data stored in an XString object are not copied on object duplication and (2) an XString object can only store a single string (see the XStringSet container for an efficient way to store a big collection of strings in a single object).

Unlike the DNAString, RNAString and AAString containers that accept only a predefined set of letters (the alphabet), a BString object can be used for storing any single string based on a single-byte character set.

In the code snippet below, x can be a single string (character vector of length 1) or an XString object.

: BString(x="", start=1, nchar=NA): Tries to convert x into a BString object by reading nchar letters starting at position start in x.

In the code snippets below, x is an XString object.

: alphabet(x): NULL for a BString object. See the corresponding man pages when x is a DNAString, RNAString or AAString object.
: length(x) or nchar(x): Get the length of an XString object, i.e., its number of letters.

In the code snippets below, x is an XString object.

: as.character(x): Converts x to a character string.
: toString(x): Equivalent to as.character(x).

In the code snippets below, x is an XString object.

x[i]: Return a new XString object made of the selected letters (subscript i must be an NA-free numeric vector specifying the positions of the letters to select). The returned object belongs to the same class as x.

Note that, unlike subseq, x[i] does copy the sequence data and therefore will be very inefficient for extracting a big number of letters (e.g. when i contains millions of positions).

In the code snippets below, e1 and e2 are XString objects.

e1 == e2: TRUE if e1 is equal to e2. FALSE otherwise.

Comparison between two XString objects of different base types (e.g. a BString object and a DNAString object) is not supported with one exception: a DNAString object and an RNAString object can be compared (see RNAString-class for more details about this).

Comparison between a BString object and a character string is also supported (see examples below).

e1 != e2: Equivalent to !(e1 == e2).

H. Pagès

subseq, letter, DNAString-class, RNAString-class, AAString-class, XStringSet-class, XStringViews-class, reverseComplement, compact, XVector-class

b <- BString("I am a BString object")
b
length(b)

## Extracting a linear subsequence:
subseq(b)
subseq(b, start=3)
subseq(b, start=-3)
subseq(b, end=-3)
subseq(b, end=-3, width=5)

## Subsetting:
b2 <- b[length(b):1]       # better done with reverse(b)

as.character(b2)

b2 == b                    # FALSE
b2 == as.character(b2)     # TRUE

## b[1:length(b)] is equal but not identical to b!
b == b[1:length(b)]        # TRUE
identical(b, 1:length(b))  # FALSE
## This is because subsetting an XString object with [ makes a copy
## of part or all its sequence data. Hence, for the resulting object,
## the internal slot containing the memory address of the sequence
## data differs from the original. This is enough for identical() to
## see the 2 objects as different.

## Compacting. As a particular type of XVector objects, XString
## objects can optionally be compacted. Compacting is done typically
## before serialization. See ?compact for more information.