Managing Gene Expression Deconvolution Algorithms

Description

gedAlgorithm provides access to the registered gene expression deconvolution algorithms.

gedAlgorithmInfo prints information about the registered gene expression deconvolution or returns -- invisibly -- the complete algorithm registry, as a registry object.

Usage

gedAlgorithm(key, error = TRUE, exact = FALSE, all = FALSE, ...)

gedAlgorithmInfo(show = TRUE, all = !show)

Arguments

key
algorithm access key, as a single character string. If missing the function returns the list of registered keys, as a character vector.
error
a logical that indicates whether an error should be thrown if the key is not found in the registry or match multiple lists. If FALSE then function returns NULL if the key is not found or NA in case of multiple matches.
exact
logical that indicates if one should use exact matching or partial matching to match the provided access key against all registered keys.
all
logical that is only used when key is missing and indicates if all registered keys should be returned, including the internal one -- whose key starts with a '.'.
...
extra arguments used internally, not to be used by the end user.
show
logical that indicates if the registry object should be printed (FALSE) or only returned invisibly (FALSE).

Examples


gedAlgorithm()
##  [1] "lsfit"       "cs-lsfit"    "qprog"       "cs-qprog"    "DSA"        
##  [6] "csSAM"       "DSection"    "ssKL"        "ssFrobenius" "meanProfile"
## [11] "deconf"
gedAlgorithm('ssKL')
## <Object of class: GEDStrategy>
## Key: ssKL
## Require: Marker
## Reference: Gaujoux et al. (2011)
gedAlgorithm('csSAM')
## <Object of class: GEDStrategy>
## Key: csSAM
## Require: Coef
## Reference: Shen-Orr et al. (2010)
# show algorithms and properties
gedAlgorithmInfo()
##                                                                                                            Description
## lsfit                                                    Partial deconvolution of proportions using least-squares fits
## cs-lsfit                                             Partial deconvolution of cell signatures using least-squares fits
## qprog                               Estimates proportions from known expression signatures using quadratic programming
## cs-qprog    Estimates constrained cell-specific signatures from proportions using quadratic programming [experimental]
## DSA                                                             Complete deconvolution using Digital Sorting Algorithm
## csSAM                                       Estimates cell/tissue specific signatures from known proportions using SAM
## DSection                                                      Estimates proportions from proportions priors using MCMC
## ssKL                                               Semi-supervised NMF algorithm for KL divergence, using marker genes
## ssFrobenius                                   Semi-supervised NMF algorithm for Euclidean distance, using marker genes
## meanProfile                                                     Compute proportion proxies as mean expression profiles
## deconf                                                  Alternate least-square NMF method, using heuristic constraints
##             Basis Coef Marker Iter     Reference
## lsfit           I    O      -    1     Abbas2009
## cs-lsfit        O    I      -    1              
## qprog           I    O      -    1      Gong2011
## cs-qprog        O    I      -    1              
## DSA             O    O      I    1     Zhong2013
## csSAM           O    I      O    1  Shen-Orr2010
## DSection        O  I/O      O  500   Erkkila2010
## ssKL            O    O      I 3000   Gaujoux2011
## ssFrobenius     O    O      I 3000   Gaujoux2011
## meanProfile     -    O      I    1              
## deconf          O    O      - 1000 Repsilber2010
class(gedAlgorithmInfo(show=FALSE))
## [1] "GEDalgorithm_registry"       "algorithms_package_registry"
## [3] "package_registry"            "registry"