somaticInteractions: Exact tests to detect mutually exclusive, co-occuring and...
In maftools: Summarize, Analyze and Visualize MAF Files

Description Usage Arguments Details Value References Examples

Performs Pair-wise Fisher's Exact test to detect mutually exclusive or co-occuring events.

somaticInteractions(
  maf,
  top = 25,
  genes = NULL,
  pvalue = c(0.05, 0.01),
  returnAll = TRUE,
  geneOrder = NULL,
  fontSize = 0.8,
  showSigSymbols = TRUE,
  showCounts = FALSE,
  countStats = "all",
  countType = "all",
  countsFontSize = 0.8,
  countsFontColor = "black",
  colPal = "BrBG",
  showSum = TRUE,
  colNC = 9,
  nShiftSymbols = 5,
  sigSymbolsSize = 2,
  sigSymbolsFontSize = 0.9,
  pvSymbols = c(46, 42),
  limitColorBreaks = TRUE
)

`maf`	an `MAF` object generated by `read.maf`
`top`	check for interactions among top 'n' number of genes. Defaults to top 25. `genes`
`genes`	List of genes among which interactions should be tested. If not provided, test will be performed between top 25 genes.
`pvalue`	Default c(0.05, 0.01) p-value threshold. You can provide two values for upper and lower threshold.
`returnAll`	If TRUE returns test statistics for all pair of tested genes. Default FALSE, returns for only genes below pvalue threshold.
`geneOrder`	Plot the results in given order. Default NULL.
`fontSize`	cex for gene names. Default 0.8
`showSigSymbols`	Default TRUE. Heighlight significant pairs
`showCounts`	Default TRUE. Include number of events in the plot
`countStats`	Default 'all'. Can be 'all' or 'sig'
`countType`	Default 'cooccur'. Can be 'all', 'cooccur', 'mutexcl'
`countsFontSize`	Default 0.8
`countsFontColor`	Default 'black'
`colPal`	colPalBrewer palettes. See RColorBrewer::display.brewer.all() for details
`showSum`	show [sum] with gene names in plot, Default TRUE
`colNC`	Number of different colors in the palette, minimum 3, default 9
`nShiftSymbols`	shift if positive shift SigSymbols by n to the left, default = 5
`sigSymbolsSize`	size of symbols in the matrix and in legend
`sigSymbolsFontSize`	size of font in legends
`pvSymbols`	vector of pch numbers for symbols of p-value for upper and lower thresholds c(upper, lower)
`limitColorBreaks`	limit color to extreme values. Default TRUE

This function and plotting is inspired from genetic interaction analysis performed in the published study combining gene expression and mutation data in MDS. See reference for details.

list of data.tables

Gerstung M, Pellagatti A, Malcovati L, et al. Combining gene mutation with gene expression data improves outcome prediction in myelodysplastic syndromes. Nature Communications. 2015;6:5901. doi:10.1038/ncomms6901.

1
2
3

laml.maf <- system.file("extdata", "tcga_laml.maf.gz", package = "maftools")
laml <- read.maf(maf = laml.maf)
somaticInteractions(maf = laml, top = 5)

-Reading
-Validating
-Silent variants: 475 
-Summarizing
-Processing clinical data
--Missing clinical data
-Finished in 0.667s elapsed (0.430s cpu) 
     gene1  gene2       pValue oddsRatio  00 11 01 10              Event
 1:   FLT3   NPM1 0.0009929836 3.7631606 125 17 16 35       Co_Occurence
 2: DNMT3A   NPM1 0.0014582861 3.7331407 128 16 17 32       Co_Occurence
 3:   IDH1 DNMT3A 0.0033807043 4.4622015 137 10 38  8       Co_Occurence
 4:   IDH2   NPM1 0.0277733049 0.0000000 140  0 33 20 Mutually_Exclusive
 5: DNMT3A   FLT3 0.0630630121 1.9514757 111 18 34 30       Co_Occurence
 6:   IDH1   NPM1 0.0914621351 2.7223023 148  6 27 12       Co_Occurence
 7:   FLT3   IDH2 0.1075065717 0.2747181 123  2 18 50 Mutually_Exclusive
 8:   IDH1   FLT3 0.4081370208 0.5158297 126  3 49 15 Mutually_Exclusive
 9:   IDH1   IDH2 0.6995972879 0.4844089 156  1 19 17 Mutually_Exclusive
10:   IDH2 DNMT3A 1.0000000000 1.0077116 130  5 43 15       Co_Occurence
            pair event_ratio
 1:   FLT3, NPM1       17/51
 2: DNMT3A, NPM1       16/49
 3: DNMT3A, IDH1       10/46
 4:   IDH2, NPM1        0/53
 5: DNMT3A, FLT3       18/64
 6:   IDH1, NPM1        6/39
 7:   FLT3, IDH2        2/68
 8:   FLT3, IDH1        3/64
 9:   IDH1, IDH2        1/36
10: DNMT3A, IDH2        5/58