[R-pkgs] The aphid package for analysis with profile hidden Markov models
shaunpwilkinson at gmail.com
Thu Aug 31 03:17:05 CEST 2017
I'm pleased to introduce a new package called ‘aphid’, for analysis with
profile hidden Markov models in R.
The package contains functions for multiple and pairwise sequence alignment
for both nucleic acids and proteins (preferably in the DNAbin or AAbin
format), model building, parameter optimization (Baum Welch and Viterbi
training), plotting, file import & export, tree-based sequence weighting,
simulation, and implementation of the forward, backward and Viterbi
algorithms for calculating sequence probabilities.
Standard HMMs are also supported, with functions for building, training,
plotting, sequence simulation, and HMM implementations of the forward,
backward & Viterbi algorithms.
The package has a wide variety of uses including database searching,
gene-finding and annotation, phylogenetic analysis and sequence
The number of dependencies has been kept to a minimum (non-base
dependencies: Rcpp, openssl & phylogram), and while aphid is designed to be
used in conjunction with the ‘ape’ package, it is not a requisite. Indeed,
the package functions will work with any standard character sequence
vectors (even non-biological ones), though at a cost of memory efficiency
when compared with ape’s raw DNAbin and AAbin objects types.
Algorithms and examples are from the essential book by Durbin et al (1998),
Biological Sequence Analysis: Probabilistic Models of Proteins and Nucleic
Acids (Cambridge University Press; ISBN 978-0521629713).
Any feedback is warmly welcomed.
Rutherford Foundation Postdoctoral Research Fellow
Victoria University of Wellington
P.O. Box 600 Wellington, New Zealand
development version: https://github.com/shaunpwilkinson/aphid
bug reports: https://github.com/shaunpwilkinson/aphid/issues
email: shaunpwilkinson at gmail.com
[[alternative HTML version deleted]]
More information about the R-packages