/fingerprint/man/bitspec.Rd

 1\name{bit.spectrum}
 2\alias{bit.spectrum}
 3\title{
 4	Generate a Bit Spectrum from a List of Fingerprints
 5}
 6\description{
 7The idea of comparing datasets using fingerprints was described in
 8Guha \& Schurer (2008). The idea is that one can summarize the dataset
 9by counting the frequency of occurrence of each bit position. The
10frequency is normalized by the number of fingerprints considered. Thus a
11collection of N fingerprints can be converted to a single vector of
12numbers highlighting the most frequent bits with respect to a given
13dataset. A plot of this vector looks like a traditional spectrum and
14hence the name.
15
16The bit spectra for two datasets (assuming that the same types of
17fingerprints have been used) allows one to compare the similarity of
18the datasets, without having to do a full pairwise similarity
19calculation. The difference between the structural features of the
20datasets can be quantified by evaluating the distance between the two
21bit spectra.
22}
23\usage{
24bit.spectrum(fplist)
25}
26\arguments{
27  \item{fplist}{
28        A list structure with each element being an object of class
29	\code{fingerprint}. These will can be constructed by hand or
30	read from disk via \code{\link{fp.read}}.
31
32	All fingerprints in the list should be of the same length.
33    }
34}
35\value{
36A numeric vector of length equal to the size of the fingerprints.
37}
38\seealso{
39    \code{\link{distance}}, \code{\link{fp.read}}
40}
41\references{
42Guha, R.; Schurer, S.; \emph{J. Comp. Aid. Molec. Des.}, \bold{2008},
43    \emph{22}, 367-384.
44}
45\keyword{programming}
46\author{Rajarshi Guha \email{rajarshi.guha@gmail.com}}