CASMAP-class {CASMAP}

R Documentation

Constructor for CASMAP class object.

Description

Constructor for CASMAP class object.

Details

Constructor for CASMAP class object, which needs the mode parameter to be set by the user. Please see the examples.

Fields

mode

Either 'regionGWAS' or 'higherOrderEpistasis'.

alpha

A numeric value setting the Family-wise Error Rate (FWER). Must be strictly between 0 and 1. Default value is 0.05.

max_comb_size

A numeric specifying the maximum length of combinations. For example, if set to 4, then only combinations of size between 1 and 4 (inclusive) will be considered. To consider combinations of arbitrary (maximal) length, use value 0, which is the default value.

Base method, for both modes

readFiles

Read the data, label and possibly covariates files. Parameters are genotype_file, for the data, phenotype_file for the labels and (optional) covariates_file for the covariates. The option plink_file_root is not supported in the current version, but will be supported in future versions.

setMode

Can set/change the mode, but note that any data files will need to read in again using the readFiles command.

setTargetFWER

Can set/change the Family-wise Error Rate (FWER). Takes a numeric parameter alpha, strictly between 0 and 1.

execute

Once the data files have been read, can execute the algorithm. Please note that, depending on the size of the data files, this could take a long time.

getSummary

Returns a data frame with a summary of the results from the execution, but not any significant regions/itemsets. See getSignificantRegions, getSignificantInteractions, and getSignificantClusterRepresentatives.

writeSummary

Directly write the information from getSummary to file.

regionGWAS Methods

getSignificantRegions

Returns a data frame with the the significant regions. Only valid when mode='regionGWAS'.

getSignificantClusterRepresentatives

Returns a data frame with the the representatives of the significant clusters. This will be a subset of the regions returned from getSignificantRegions. Only valid when mode='regionGWAS'.

writeSignificantRegions

Writes the data from getSignificantRegions to file, which must be specified in the parameter path. Only valid when mode='regionGWAS'.

writeSignificantClusterRepresentatives

Writes the data from getSignificantClusterRepresentatives to file, which must be specified in the parameter path. Only valid when mode='regionGWAS'.

higherOrderEpistasis Methods

getSignificantInteractions

Returns the frame from getSignificantInteractions to file, which must be specified in the parameter path. Only valid when mode='higherOrderEpistasis'.

writeSignificantInteractions

Writes a data frame with the significant interactions. Only valid when mode='higherOrderEpistasis'.

References

A. Terada, M. Okada-Hatakeyama, K. Tsuda and J. Sese Statistical significance of combinatorial regulations, Proceedings of the National Academy of Sciences (2013) 110 (32): 12996-13001

F. Llinares-Lopez, D. G. Grimm, D. Bodenham, U. Gieraths, M. Sugiyama, B. Rowan and K. Borgwardt, Genome-wide detection of intervals of genetic heterogeneity associated with complex traits, ISMB 2015, Bioinformatics (2015) 31 (12): i240-i249

L. Papaxanthos, F. Llinares-Lopez, D. Bodenham, K .Borgwardt, Finding significant combinations of features in the presence of categorical covariates, Advances in Neural Information Processing Systems 29 (NIPS 2016), 2271-2279.

F. Llinares-Lopez, L. Papaxanthos, D. Bodenham, D. Roqueiro and K .Borgwardt, Genome-wide genetic heterogeneity discovery with categorical covariates. Bioinformatics 2017, 33 (12): 1820-1828.

Examples

## An example using the "regionGWAS" mode
fastcmh <- CASMAP(mode="regionGWAS")      # initialise object

datafile <- getExampleDataFilename()      # file name of example data
labelsfile <- getExampleLabelsFilename()  # file name of example labels
covfile <- getExampleCovariatesFilename() # file name of example covariates 

# read the data, labels and covariate files
fastcmh$readFiles(genotype_file=getExampleDataFilename(),
                  phenotype_file=getExampleLabelsFilename(), 
                  covariate_file=getExampleCovariatesFilename() )

# execute the algorithm (this may take some time)
fastcmh$execute()

#get the summary results
summary_results <- fastcmh$getSummary()

#get the significant regions
sig_regions <- fastcmh$getSignificantRegions()

#get the clustered representatives for the significant regions
sig_cluster_rep <- fastcmh$getSignificantClusterRepresentatives()


## Another example of regionGWAS
fais <- CASMAP(mode="regionGWAS")      # initialise object

# read the data and labels, but no covariates
fastcmh$readFiles(genotype_file=getExampleDataFilename(),
                  phenotype_file=getExampleLabelsFilename())


## Another example, doing higher order epistasis search
facs <- CASMAP(mode="higherOrderEpistasis")      # initialise object

---