ppi.infer.human {PPInfer} | R Documentation |
This function is designed for human protein-protein interaction from STRING database. Default format is 'hgnc'. The number of proteins is 10 in default. Note that the number of proteins used as a target may be different from the number of proteins in the input since mapping between formats is not always one-to-one in getBM.
ppi.infer.human(target, kernel, top = 10, classifier = net.infer, input = "hgnc_symbol", output = "hgnc_symbol", ...)
target |
set of interesting proteins or target class |
kernel |
the regularized Laplacian matrix for a graph |
top |
number of top proteins most closely related to target class (default: 10) |
classifier |
net.infer or net.infer.ST (default: net.infer) |
input |
input format |
output |
output format |
... |
additional parameters for the chosen classifier |
list |
list of a target class used in the model |
error |
training error |
CVerror |
cross validation error, (when cross > 0 in net.infer) |
top |
top proteins |
score |
decision values for top proteins |
Dongmin Jung, Xijin Ge
net.infer, net.infer.ST, getBM
# example 1 string.db.9606 <- STRINGdb$new(version = '11', species = 9606, score_threshold = 999) string.db.9606.graph <- string.db.9606$get_graph() K.9606 <- net.kernel(string.db.9606.graph) rownames(K.9606) <- substring(rownames(K.9606), 6) colnames(K.9606) <- substring(colnames(K.9606), 6) target <- colnames(K.9606)[1:100] infer.human <- ppi.infer.human(target, K.9606, input = "ensembl_peptide_id") ## Not run: # example 2 library(graph) data(apopGraph) target <- nodes(apopGraph) apoptosis.infer <- ppi.infer.human(target, K.9606, 100) # example 3 library(KEGGgraph) library(KEGG.db) pName <- "p53 signaling pathway" pId <- mget(pName, KEGGPATHNAME2ID)[[1]] getKGMLurl(pId, organism = "hsa") p53 <- system.file("extdata/hsa04115.xml", package="KEGGgraph") p53graph <- parseKGML2Graph(p53,expandGenes=TRUE) entrez <- translateKEGGID2GeneID(nodes(p53graph)) ensembl <- useMart("ensembl") human.ensembl <- useDataset("hsapiens_gene_ensembl",mart=ensembl) target <- getBM(attributes=c('entrezgene', 'hgnc_symbol'), filter = 'entrezgene', values = entrez, mart = human.ensembl)[,2] p53.infer <- ppi.infer.human(target, K.9606, 100) ## End(Not run)