RAVA_FIRST: RAVA-FIRST: RAre Variant Association using...

Description Usage Arguments Details Value Source See Also Examples

Description

Analyse rare variants using the RAVA-FIRST approach based on CADD scores to group and filter rare variants

Usage

1
2
3
4
5
6
RAVA.FIRST(x, variant.scores = NULL, ref.level, 
           filter=c("whole", "controls", "any"), 
           maf.threshold=0.01, min.nb.snps = 2, 
           min.cumulative.maf = NULL, group = NULL, 
           cores = 10, burden = TRUE, H0.burden, burden.parameters, 
           SKAT = TRUE, H0.SKAT, SKAT.parameters, verbose = TRUE)

Arguments

x

A bed.matrix

variant.scores

A dataframe containing the adjusted CADD scores of the variants (Optional, useful to gain in computation time if the adjusted CADD scores of variants in the study are available)

ref.level

The level corresponding to the controls group, only needed if filter=="controls"

filter

On which group the MAF filter will be applied

maf.threshold

The MAF threshold used to define a rare variant, set at 0.01 by default

min.nb.snps

The minimum number of variants needed to keep a CADD region, set at 2 by default

min.cumulative.maf

The minimum cumulative maf of variants needed to keep a CADD region

group

A factor indicating the group of each individual, only needed if filter = "controls" or filter = "any". If missing, x@ped$pheno is taken

cores

How many cores to use, set at 10 by default

burden

Whether to compute the burden test

H0.burden

A list returned from NullObject.parameters with RVAT="burden"

burden.parameters

A list containing the parameters to use by burden.subscores for the burden analysis ('burden.function' and 'get.effect.size')

SKAT

Whether to compute SKAT

H0.SKAT

A list returned from NullObject.parameters with RVAT="SKAT"

SKAT.parameters

A list containing the parameters to use by SKAT ('get.moments', 'estimation.pvalue', 'params.sampling', 'debug')

verbose

Whether to display information about the function actions

Details

Rare variants are analysed using the 'RAVA-FIRST' strategy composed of three steps: - Rare variants are grouped in 'CADD regions' defined from the CADD scores of variants observed in GnomAD. - Rare variant are selected within each CADD region based on an adjusted CADD score using a region-specific threshold corresponding to the median of scores observed in GnomAD in each region. - Burden analysis is performed by integrating sub-scores for the coding, regulatory and intergenic categories within each CADD region. For SKAT analysis, a test for each CADD region is performed.

RAVA.FIRST() is based on the functions set.CADDregions, filter.adjustedCADD, burden.subscores and SKAT. Please refer to these functions for more information. Especially, refer to the functions burden.subscores and SKAT to get more information about what is need in burden.parameters and SKAT.parameters.

It is recommended to use this function chromosome by chromosome for large datasets.

Value

A list containing the results for the burden analysis ('burden') and the results for the SKAT analysis ('SKAT'), along with information about CADD regions (positions, type of genomic categories overlapped by each region and median of adjusted CADD scores).

Source

https://lysine.univ-brest.fr/RAVA-FIRST/

See Also

set.CADDregions, filter.adjustedCADD, burden.subscores, SKAT

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
#Import 1000Genome data from region around LCT gene
#x <- as.bed.matrix(x=LCT.matrix.bed, fam=LCT.matrix.fam, bim=LCT.snps)

#Add population
#x@ped[,c("pop", "superpop")] <- LCT.matrix.pop1000G[,c("population", "super.population")]

#Select EUR superpopulation
#x <- select.inds(x, superpop=="EUR")
#x@ped$pop <- droplevels(x@ped$pop)

#Perform RAVA-FIRST with burden analysis
#H0.burden <- NullObject.parameters(pheno = x@ped$pop, ref.level = "CEU",
#                                   RVAT = "burden", pheno.type = "categorial")
#res.burden <- RAVA.FIRST(x, maf.threshold = 0.05,
#                         H0.burden = H0.burden, SKAT = F)

Ravages documentation built on Oct. 19, 2021, 5:07 p.m.