| Type: | Package |
| Title: | Pathway Analysis Methods for Genomewide Association Data |
| Version: | 2.0 |
| Date: | 2015-12-07 |
| Author: | Marina Evangelou |
| Maintainer: | Marina Evangelou <m.evangelou@ic.ac.uk> |
| Description: | Bayesian hierarchical methods for pathway analysis of genomewide association data: Normal/Bayes factors and Sparse Normal/Adaptive lasso. The Frequentist Fisher's product method is included as well. |
| License: | GPL-2 | GPL-3 [expanded from: GPL (≥ 2)] |
| Depends: | lars, foreach, mnormt |
| Packaged: | 2015-12-08 12:35:35 UTC; marinaevangelou |
| NeedsCompilation: | no |
| Repository: | CRAN |
| Date/Publication: | 2015-12-08 14:53:45 |
Pathway Analysis Methods for Genomewide Association Data
Description
Bayesian hierarchical methods for pathway analysis of genomewide association data: Normal/Bayes factors and Sparse Normal/Adaptive lasso. The Frequentist Fisher's method is included as well.
Details
| Package: | PAGWAS |
| Type: | Package |
| Version: | 2.0 |
| Date: | 2015-12-02 |
| License: | GPL (>=2) |
| LazyLoad: | yes |
Author(s)
Marina Evangelou Maintainer: Marina Evangelou <m.evangelou@ic.ac.ukt>
References
Evangelou, M., Dudbridge, F., Wernisch, L. (2014). Two novel pathway analysis methods based on a hierarchical model. Bioinformatics 30(5), 690 - 697
Evangelou, M., Rendon, A., Ouhewand, W. H., Wernisch, L., Dudbridge, F. (2012) Comparison of methods for competitive tests of pathway analysis. Plos One 7(7): e41018
Calculates the Fisher's method p-value for each tested pathway
Description
Calculates the Fisher's method p-value for a set of p-values. It returns both the p-value and the test statistic value of the Fisher's product method.
Usage
FM.chi.pvalue(x)
Arguments
x |
A vector of p-values. These p-values can be either gene or SNP p-values of a tested pathway |
Value
FMstatistic |
Fisher's product method test statistic |
FMpvalue |
Fisher's method p-value, computed using the exact distribution of the Fisher's method test statistic which is a Chi^2 distribution with degrees of freedom twice the size of vector x |
References
Evangelou M, Rendon A, Ouwehand WH, Wernisch L, Dudbridge F (2012) Comparison of Methods for Competitive Tests of Pathway Analysis. PLoS ONE 7(7): e41018. doi:10.1371/journal.pone.0041018
See Also
Examples
FM.chi.pvalue(x=c(0.05,0.1))Normal/Bayes factors method for finding associated pathways
Description
A vector of the computed Bayes factors for the tested pathways.
Usage
NBF(y, G, P, a, b, s2, nu)
Arguments
y |
Response vector of length N |
G |
Genotype matrix, with N rows and L columns (number of tested SNPs) |
P |
Pathway matrix, with L columns and M columns (number of tested pathways) |
a |
Hyper-parameter of the variance assumed for the integrated out SNP effects |
b |
Hyper-parameter of the variance assumed for the pathway effects |
s2 |
Hyper-parameter of the Inverse-Chi-squared distribution assumed for the variance of the response vector |
nu |
Hyper-parameter of the Inverse-Chi-squared distribution assumed for the variance of the response vector |
Value
A vector of the computed Bayes factors of the same length as the number of tested pathways
References
Evangelou, M., Dudbridge, F., Wernisch, L. (2014). Two novel pathway analysis methods based on a hierarchical model. Bioinformatics, 30(5), 690 - 697.
Examples
## Not run:
data(genotypes)
G=genotypes
data(pathways)
data(SNPs)
data(genes)
snps.genes=snps.to.genes(SNPs,genes,distance=0)
snps.paths=snps.to.pathways(pathways,snps.genes)
P=create.pathway.df(G,snps.paths)
y=rnorm(nrow(G),mean=0,sd=10)
NBF(y,G,P,a,b,s2,nu)
## End(Not run)
Sparse Normal/Adaptive lasso method for finding associated pathways
Description
Sparse Normal/Adaptive lasso method applied for finding the associated pathways. The iterative algorithm suggested by Wipf and Nagarajan (2008) is applied. A vector equal to the number of tested pathways is returned, the zero entries of the vector correspond to the pathways that are not associated. The posterior estimates of the beta coefficients are also returned as they are described by Wipf and Nagarajan (2008).
Usage
SNAL(y, G, P, a, s2)
Arguments
y |
Response vector of length N |
G |
Genotype matrix, with N rows and L columns (number of tested SNPs) |
P |
Pathway matrix, with L columns and M columns (number of tested pathways) |
a |
Hyper-parameter of the variance assumed for the integrated out SNP effects |
s2 |
Variance assumed for the response variable, the tuning parameter of adaptive lasso |
Value
gamma.star |
Estimates of gamma hyper-parameters |
ARD |
Posterior estimates of beta coefficients |
References
Evangelou, M., Dudbridge, F., Wernisch, L. (2014). Two novel pathway analysis methods based on a hierarchical model. Bioinformatics, 30(5), 690 - 697.
Wipf, D. and Nagarajan, S. (2008). A new view of automatic relevance determination. Advances in Neural Information Processing Systems, 20
See Also
Examples
## Not run:
data(genotypes)
G=genotypes
data(pathways)
data(SNPs)
data(genes)
snps.genes=snps.to.genes(SNPs,genes,distance=0)
snps.paths=snps.to.pathways(pathways,snps.genes)
P=create.pathway.df(G,snps.paths)
y=rnorm(nrow(G),mean=0,sd=10)
SNAL(y,G,P,a,s2)
## End(Not run)
Sparse Normal/Adaptive lasso method for finding associated variables. The SNAL method is applied to the linear regression Y= Phi beta + epsilon
Description
For more details please read SNAL.
Usage
SNAL.calculation(Y, Phi, s2)
Arguments
Y |
Response vector of length N |
Phi |
Design matrix, with N rows and M columns (number of tested variables) |
s2 |
Variance assumed for the response variable, the tuning parameter of the adaptive lasso problem |
Value
gamma.star |
Estimates of gamma hyper-parameters |
ARD |
Posterior estimates of beta coefficients |
References
Evangelou, M., Dudbridge, F., Wernisch, L. (2014). Two novel pathway analysis methods based on a hierarchical model. Bioinformatics, 30(5), 690 - 697
Wipf, D. and Nagarajan, S. (2008). A new view of automatic relevance determination. Advances in Neural Information Processing Systems, 20
See Also
Examples
## Not run: SNAL.calculation(Y,Phi,s2=0.5)
A data frame of 100 artificial SNPs with their chromosomes and positions on the genome
Description
A data frame with 100 rows and 3 columns.
Usage
data(SNPs)Format
Column names:
NameSNP name
PositionPosition of SNP on the genome
ChrChromosome of the SNP
See Also
Examples
data(SNPs)
print(SNPs[1:5,])
Creates a pathway data frame
Description
Returns a data frame with L rows and M columns. L is the number of SNPs in the genotypes data frame and M is the number of tested pathways.
Usage
create.pathway.df(genotypes,snps.paths)
Arguments
genotypes |
Genotype matrix, with L SNPs (columns) and N individuals (rows) |
snps.paths |
A list with entries the SNP members of each pathway. The size of the list is M |
Value
A data frame with columns equal to the number of pathways in the pathway.snps list and rows equal to the number of tested SNPs
See Also
SNPs, genes, snps.to.pathways
snps.to.genes
Examples
data(SNPs)
data(genes)
data(pathways)
data(genotypes)
snps.genes <- snps.to.genes(snp.info=SNPs,gene.info=genes, distance=0)
pathway.snps <- snps.to.pathways(pathways,snps.genes)
P <- create.pathway.df(genotypes=genotypes,snps.paths=pathway.snps)
A data frame of 20 artificial genes with their chromosomes and positions on the genome
Description
A data frame with 20 rows and 4 columns.
Usage
data(genes)Format
Column names:
NameName of gene
StartStart position of gene on the genome
EndEnd position of gene on the genome
ChrChromosome of gene
See Also
Examples
data(genes)
print(genes[1:5,])
Genotypes for 100 SNPs and 75 individuals
Description
A data frame with 75 rows (individuals) and 100 columns (SNPs). The entries of the genotype matrix are 0, 1 and 2. There are no missing values.
Usage
data(genotypes)See Also
Examples
data(genotypes)
A list of 2 pathways with their gene members
Description
A list of two pathways. The gene members of each pathway are given.
Usage
data(pathways)See Also
Examples
data(pathways)
Assigns SNPs to genes
Description
Assigns SNPs to genes based on their physical distance.
Usage
snps.to.genes(snp.info, gene.info, distance)
Arguments
snp.info |
A data frame with 3 columns with names: Name, Position and Chr that correspond to the SNP name, its position on the genome and its chromosome, respectively |
gene.info |
A data frame with 4 columns with names: Name, Start, End and Chr that correspond to the gene name, start and end positions on the genome and its chromosome, respectively |
distance |
A number that corresponds to the distance below and above the Start and End positions of the gene that all SNPs in that region should be assigned to the gene |
Value
A list of the same size as the number of genes of the gene.info data frame. The names of the SNPs assigned to each gene are returned
See Also
Examples
data(SNPs)
data(genes)
snps.to.genes(snp.info=SNPs,gene.info=genes,distance=50)
Assigns SNPs to pathways
Description
Assigns SNPs to pathways, using the pathway gene members and the SNPs assigned to each gene.
Usage
snps.to.pathways(pathways,gene.snps)
Arguments
pathways |
A list of pathways with their gene members |
gene.snps |
A list of genes with the SNPs assigned to them according to their physical distance on the genome |
Value
A list of the same size as the number of pathways in the pathway list. The names of the SNPs assigned to each pathway are returned. Empty pathways are also returned.
See Also
Examples
data(SNPs)
data(genes)
data(pathways)
snps.genes <- snps.to.genes(snp.info=SNPs,gene.info=genes, distance=50)
pathway.snps <- snps.to.pathways(pathways,snps.genes)