analyzeFI: Functional interaction analysis of AS events

In ASpediaFI: ASpedia-FI: Functional Interaction Analysis of Alternative Splicing Events

Description

Analyze functional interactions of AS events using Discriminative Random Walk with Restart (DRaWR). It runs a DRaWR on a heterogeneous network containing genes, AS events, and pathways. It then performs GSEA on gene sets related to query genes.

Usage

analyzeFI(object, query, expr, ppi = NULL, pathways = NULL,
    restart = 0.7, num.folds = 5, num.feats = 100, low.expr = 1,
    low.var = NULL, prop.na = 0.05, prop.extreme = 1,
    cor.threshold = 0.3)

Arguments

object	Object of class ASpediaFI
query	a character vector or a data frame containing query genes
expr	a SummarizedExperiment object or matrix containing gene expression profiles (FPKM)
ppi	an igraph object containing known interactions between genes. If NULL, an igraph object containing human gene-gene interactions will be used.
pathways	a GMT file or a named list of pathway gene sets. If NULL, a combined list of HALLMARK, KEGG, and REACTOME pathway gene sets will be used.
restart	a restart probability
num.folds	the number of folds for cross-validation
num.feats	the number of feature nodes to be retained in the final subnetwork
low.expr	Genes with mean expression below low.expr are excluded. AS events for corresponding genes are also excluded.
low.var	AS events with variance below low.var are excluded. If NULL, top 10,000 variable events are used for analysis.
prop.na	AS events with the higher proportion of missing values than prop.na are excluded.
prop.extreme	AS events with the higher proportion of extreme values (0 or 1) than prop.extreme are excluded.
cor.threshold	a pair of AS event and gene with Spearman's correlation greather than cor.threshold are connected in a heterogeneous network.

Value

ASpediaFI object with results of functional interaction analysis

References

Blatti, C. et al. (2016). Characterizing gene sets using discrminative random walks with restart on heterogeneous biological networks. Bioinformatics, 32.

Examples

library(limma)
data(GSE114922.fpkm)
data(GSE114922.psi)
design <- cbind(WT = 1, MvsW = colData(GSE114922.psi)$condition == 'MUT')
fit <- lmFit(log2(GSE114922.fpkm + 1), design = design)
fit <- eBayes(fit, trend = TRUE)
tt <- topTable(fit, number = Inf, coef = 'MvsW')
query <- rownames(tt[tt$logFC > 1 & tt$P.Value < 0.1, ])
head(query)
## Not run: 
GSE114922.ASpediaFI <- analyzeFI(
    GSE114922.ASpediaFI, query,
    GSE114922.fpkm
)

## End(Not run)

ASpediaFI documentation built on Nov. 8, 2020, 8:13 p.m.