GSNA (version 0.1.4.2)

gsn_default_distance: gsn_default_distance, gsn_distances, pw_id_col, pw_stat_col, pw_sig_order, pw_stat_col_2, pw_sig_order_2, pw_n_col, pw_type

Description

Get and set values for GSNData internal fields. When evaluated directly or on the right side of an <- assignment, these functions retrieve stored values. When evaluated on the left side of a <- assignment, they set the relevant field values.

Usage

gsn_default_distance(object)
gsn_default_distance(object) <- value
gsn_distances(object)
pw_id_col(object)
pw_id_col(object) <- value
pw_stat_col(object)
pw_stat_col(object) <- value
pw_stat_col_2(object)
pw_stat_col_2(object) <- value
pw_sig_order(object)
pw_sig_order(object) <- value
pw_sig_order_2(object)
pw_sig_order_2(object) <- value
pw_n_col(object)
pw_n_col(object) <- value
pw_type(object)
pw_type(object) <- value

Value

For get versions of the functions, evaluated alone or on the right side of a <-

assignment, the values stored in the relevant fields are returned, generally as character vectors of length 1 (or NULL), except for gsn_distances which may return character vectors of varying length. For set versions of the functions, for which the function call is on the left side of a <-

assignment, a copy of the GSNData object with the specified field set is returned. See Details.)

Arguments

object

A GSNData object.

value

A valid value for the field to be set. (see Details.).

Details

gsn_default_distance()

Gets and sets the value stored in the obect$default_distance field of a GSNData object. When a GSNData object contains distance matrices of multiple types (e.g. single-tail log fisher & Jaccard, the defailt_distance field tells GSNA which distance to use for network paring, subnet assignment, plotting, etc. When setting the value, it must be a character vector of length 1 that contains the name of a valid distance metric for which there exists a distance matrix in the GSNData object, or else an error will be raised.

gsn_distances()

Returns a character vector containing the names of the distances for which there are distance matrices in a GSNData object.

pw_id_col()

Gets and sets the value stored in the obect$pathways$id_col field. This is the field that determines which column in a pathways data.frame corresponds to a gene set identifier used in a gene set collection list of vectors, or a tmod or tmodGS object. When setting the value, this the function checks that the value is a valid column in obect$pathways$data.

pw_stat_col()

Gets and sets the value stored in the obect$pathways$stat_col field. This is the field that determines which column in a pathways data.frame corresponds to a significance statistic of interest. When setting the value, this the function checks that the value is a valid column in obect$pathways$data.

pw_sig_order()

Gets and sets the value stored in the obect$pathways$sig_order field. This is the field that states the behavior of the significance value in the obect$pathways$sig_order field, specifically whether low or high values are significant. This may be either 'loToHi' or 'hiToLo'. (Other types of statistics are possible, for example statistics with significant of high or low absolute values. We hope to add support for such statistics in the future.)

pw_stat_col_2()

Gets and sets the value stored in the obect$pathways$stat_col_2 field. For two-channel GSNA analysis, this is the field that determines which column in a pathways data.frame corresponds to the second significance statistic of interest. When setting the value, this the function checks that the value is a valid column in obect$pathways$data.

pw_sig_order_2()

Gets and sets the value stored in the obect$pathways$sig_order_2 field. For a two-channel GSNA analysis, this is the field that states the behavior of the significance value in the obect$pathways$sig_order_2 field, specifically whether low or high values are significant. This may be either 'loToHi' or 'hiToLo'. (Other types of statistics are possible, for example statistics with significant of high or low absolute values. We hope to add support for such statistics in the future.)

pw_n_col()

Gets and sets the value stored in the obect$pathways$n_col field. This is the field that determines which column in a pathways data.frame corresponds to a gene set size or gene set effective size. When setting the value, this the function checks that the value is a valid column in obect$pathways$data.

pw_type()

Gets and sets the value stored in the obect$pathways$type field. This is the field that describes what kind of pathways data are stored in the object, e.g. 'gsea' or 'cerno'. If there is no current pathways data in the object an error will be raised.

Examples

# These examples require some setup.
#
# First, we will generate a gene set network from CERNO example
# data, containing multiple distance metrics, as well as pathways
# data. We begin by subsetting the CERNO data for significant results:
sig_pathways.cerno <- subset( Bai_CiHep_DN.cerno, adj.P.Val <= 0.05 )

# Now create a gene set collection containing just the gene sets
# with significant CERNO results, by subsetting Bai_gsc.tmod using
# the gene set IDs as keys:
sig_pathways.tmod <- Bai_gsc.tmod[sig_pathways.cerno$ID]

# And obtain a background gene set from differential expression data:
background_genes <- toupper( rownames( Bai_CiHep_v_Fib2.de ) )

# Create a GSNData object containing Jaccard indices:
sig_pathways.GSN <-
   buildGeneSetNetworkJaccard(geneSetCollection = sig_pathways.tmod,
                              ref.background = background_genes )

# Within the same object, add an 'stlf' (Single Tail Log Fisher)
# distance matrix:
sig_pathways.GSN <-
   buildGeneSetNetworkSTLF( object = sig_pathways.GSN )

# Now import the CERNO data:
sig_pathways.GSN <- gsnAddPathwaysData( sig_pathways.GSN,
                                        pathways_data = sig_pathways.cerno )

# Use gsn_distances() to see what distances are stored in the
# GSNData object:
gsn_distances( sig_pathways.GSN )
# Should return: "jaccard" "stlf"

# See what the default distance is:
gsn_default_distance( sig_pathways.GSN )
# Returns: "stlf". Let's change the default distannce
# to "jaccard":
gsn_default_distance( sig_pathways.GSN ) <- "jaccard"

# Let's examine what the ID column is:
pw_id_col( sig_pathways.GSN )
# Returns: "ID"
pw_id_col( sig_pathways.GSN ) <- "ID"
# This is equivalent to the following code. When
# invoked on the left side of an assignment, R uses
# *syntactic sugar* to comvert the call to:
sig_pathways.GSN <- `pw_id_col<-`( object = sig_pathways.GSN,
                                   value = "ID" )

# On the other hand, the following returns an error
# because there is no column in the pathways dataframe
# named "invalid.name":
 class( try( pw_id_col( sig_pathways.GSN ) <- "invalid.name" ) )
 # "try-error"



# Likewise we can get and set the value of stat_col
# and sig_order:
pw_stat_col(sig_pathways.GSN )
# Returns "adj.P.Val". Let's set it to "AUC"
pw_stat_col(sig_pathways.GSN) <- "AUC"
# And likewise, sig_order:
pw_sig_order(sig_pathways.GSN) # "loToHi"
pw_sig_order(sig_pathways.GSN) <- "hiToLo"


# For 2-channel GSNA analyses, we can set the values
# of stat_col_2 and sig_order_2:
pw_stat_col_2(sig_pathways.GSN )
# Returns NULL. Let's set it to "P.Value"
pw_stat_col_2(sig_pathways.GSN) <- "P.Value"
# And likewise, sig_order:
pw_sig_order_2(sig_pathways.GSN) # NULL
pw_sig_order(sig_pathways.GSN) <- "loToHi"

# pw_n_col() works the same way to set n_col:
pw_n_col(sig_pathways.GSN) # "N1"
pw_n_col(sig_pathways.GSN) <- "N1"

# And also, pw_type()
pw_type(sig_pathways.GSN) # "cerno"
# For setting via pw_type, the value is not
# currently checked, since pathways data may
# be of many types:
pw_type(sig_pathways.GSN) <- "other"

pw_type(sig_pathways.GSN) # "other"