GSNA (version 0.1.4.2)

yassifyPathways: yassifyPathways

Description

Takes a data.frame and outputs an attractively formatted HTML table widget for reports via the using the DT and data.table package. Optionally, the user can specify, via the n parameter, the number of rows to display in the HTML table. Optionally, IDs in specific columns can be mapped to URLs as links.

Usage

yassifyPathways(
  pathways,
  n = NULL,
  url_map_list = list(),
  url_map_by_words_list = list(),
  min_decimal = 5e-04,
  quiet = TRUE,
  table_row_colors = c(`1` = "#EEF", `2` = "#FFD"),
  ...
)

Value

An attractive HTML table widget, optionally with unique IDs represented as links.

Arguments

pathways

A data.frame containing pathways data.

n

(optional) The number of rows that the user wishes to display. This defaults to the total number of rows.

url_map_list

(optional) A list of vectors containing unique IDs and their corresponding URLs as name:value pairs. In the enclosing list, the element names are the names of the columns in the data.frame containing the fields needing to be converted to URL links.

url_map_by_words_list

(optional) Similar to url_map_list, except that instead of mapping the full value of a text field, the function looks for occurrences of key values within the text using gsub() to substitute a URL link tag. This allows fields containing multiple IDs to be converted to a group of URL links.

min_decimal

(optional) The minimal value for decimal format. Below this, scientific notation is used (default 0.0005).

quiet

(optional) If FALSE, this tells the function to emit warnings when an identifier term has no matching URL. By default, this value is TRUE, suppressing this behavior.

table_row_colors

(optional) This argument specifies the row background colors used to contrast different subnets. (default: c("1"="#EEF","2"="#FFD"), pale blue and pale yellow)

...

Additional arguments passed to DT::datatable.

Examples

# The sample data object Bai_CiKrt_DN.cerno contains MSigDB
# systematic names as gene set identifiers in its ID column
# that we can map to URLs on MSigDB's website using the
# 'systematicName' URL parameter:
msig_url <- "http://www.gsea-msigdb.org/gsea/msigdb/geneset_page.jsp"
id2url <- with( Bai_CiKrt_DN.cerno,
                structure(paste0( msig_url, "?systematicName=", ID),
                          names = ID
                         )
              )

# NOTE: In GSEA data sets against MSigDB,
# MSigDB STANDARD_NAMES (e.g. "GO_RESPONSE_TO_GLUCAGON")
# are often present in the pathways data instead of
# systematic name identifiers. They can be linked to URLs
# using the 'geneSetName' parameter, as follows:
#  sn2url <-
#   with( Bai_CiKrt_DN.gsea,
#       structure( paste0(msig_url, "?geneSetName=", STANDARD_NAME),
#                  names = STANDARD_NAME
#                )
#       )

# The named vector id2url now contains URLs for MSigDB
# gene sets, names with the gene set ID. By passing a
# list containing the id2url named as the column we
# wish to map to a URL, we can have yassifyPathways
# generate an HTML table with links for the gene set IDs.
yassifyPathways( Bai_CiKrt_DN.cerno,
                 n = 200,
                 url_map_list = list(ID = id2url) )
# Here the 'n = 200' argument tells the function to
# generate an HTML table with just the first 200 results,
# and the 'url_map_list = list(ID=id2url)' tells the
# function to link the ID column of Bai_CiKrt_DN.cerno
# to the mapped URLs in the 'id2url' vector. In this case
# the entire ID field is mapped, but if we want to map
# in a word-based fashion, for example when a column
# may contain multiple IDs per row (eg "M40804, M40775" ),
# then the 'url_map_by_words_list = list(ID = id2url)'
# argument works:
yassifyPathways( Bai_CiKrt_DN.cerno,
                 n = 200,
                 url_map_by_words_list = list(ID = id2url) )
# The url_map_list_by_words argument will work in mos
# cases where url_map_list does, so may be fine to use
# generally, but it is less efficient and my sometimes be
# slower.