Skip to content

Genetic association tests using SAIGE

Jump to bottom
weizhouUMICH edited this page Apr 18, 2019 · 53 revisions

Overview

SAIGE version > 0.35.6 supports single-variant association tests, gene-based tests (SKAT-O, SKAT, BURDEN), as well as conditional analysis based on summary statistics for both single variant and gene-based tests.

This tutorial will walk you through

  1. Installing SAIGE
  2. Running SAIGE for
  • Single-variant association tests
  • Gene-based tests
  • Conditional analysis

Workflow of these tests are almost the same. They use the same step 1 and step 2 functions of SAIGE. We are going to use the example data in the extdata folder in the github page.

Installing SAIGE

SAIGE can be installed in 4 ways. The first 3 ways require installing dependencies first.

  • R-3.5.1, gcc >= 5.5.0, cmake 3.8.1

  • R packages: Rcpp, RcppArmadillo, RcppParallel, data.table, SPAtest, RcppEigen, Matrix, methods, optparse

  • /extdata/install_packages.R can be used to install the R packages

  1. The binary install file can be downloaded from SAIGE releases or from the master branch. Then using the following command line to install SAIGE
R CMD INSTALL SAIGE_XX_R_x86_64-pc-linux-gnu.tar.gz
  1. Installing using the R library devtools from github
devtools::install_github("weizhouUMICH/SAIGE")
  1. Installing from the source code.
src_branch=master
repo_src_url=https://github.com/weizhouUMICH/SAIGE
git clone --depth 1 -b $src_branch $repo_src_url

R CMD INSTALL SAIGE
  1. Using a docker image (Thanks to Juha Karjalainen for sharing the Dockerfile). The docker image can be pulled
docker pull wzhou88/saige:0.35.8.2

Dockerfile for creating a docker image for SAIGE

Functions can be called

step1_fitNULLGLMM.R --help
step2_SPAtests.R --help
createSparseGRM.R --help

Flowchart

SAIGE and SAIGE-GENE can work for both binary and quantitative traits.

SAIGE

2 steps in SAIGE

SAIGE-GENE

2 steps in SAIGE-GENE

Running SAIGE and SAIGE-GENE

Scripts are in the folder extdata/ and command lines are in the file cmd.sh

cd /extdata/
less -S cmd.sh

Input files are in the folder extdata/input
Output files are in the folder extdata/output

To obtain help information of the scripts that call functions in the SAIGE library

Rscript createSparseGRM.R --help
Rscript step1_fitNULLGLMM.R --help
Rscript step2_SPAtests.R --help

To obtain help information of functions in the SAIGE library

#open R
R
library(SAIGE)
?createSparseGRM
?fitNULLGLMM
?SPAGMMATtest

Single variant association tests

For single-variant association tests, sparse GRM and categorical variance ratios are NOT needed. Randomly selected markers with MAC >= 20 are used to estimate the variance ratio

Step 1: fitting the null logistic/linear mixed model

  • For binary traits, a null logistic mixed model will be fitted (--traitType=binary).
  • For quantitative traits, a null linear mixed model will be fitted (--traitType=quantitative) and needs to be inverse normalized (--invNormalize=TRUE) 
#check the help info for step 1
Rscript step1_fitNULLGLMM.R --help

#For Binary traits:
Rscript step1_fitNULLGLMM.R     \
        --plinkFile=./input/nfam_100_nindep_0_step1_includeMoreRareVariants_poly \
        --phenoFile=./input/pheno_1000samples.txt_withdosages_withBothTraitTypes.txt \
        --phenoCol=y_binary \
        --covarColList=x1,x2 \
        --sampleIDColinphenoFile=IID \
        --traitType=binary        \
        --outputPrefix=./output/example_binary \
        --nThreads=4 \
        --LOCO=FALSE

#For quantitative traits, if not normally distributed, inverse normalization needs to be specified to be TRUE --invNormalize=TRUE
Rscript step1_fitNULLGLMM.R     \
        --plinkFile=./input/nfam_100_nindep_0_step1_includeMoreRareVariants_poly \
        --phenoFile=./input/pheno_1000samples.txt_withdosages_withBothTraitTypes.txt \
        --phenoCol=y_quantitative \
        --covarColList=x1,x2 \
        --sampleIDColinphenoFile=IID \
        --traitType=quantitative       \
	--invNormalize=TRUE	\
        --outputPrefix=./output/example_quantitative \
        --nThreads=4 \
        --LOCO=FALSE	\
	--tauInit=1,0
Input files
  1. Genotype file for constructing the genetic relationship matrix (GRM)
    SAIGE takes the PLINK binary file for the genotypes and assumes the file prefix is the same one for .bed, .bim. and .fam
./input/nfam_100_nindep_0_step1_includeMoreRareVariants_poly.bed
./input/nfam_100_nindep_0_step1_includeMoreRareVariants_poly.bim
./input/nfam_100_nindep_0_step1_includeMoreRareVariants_poly.fam
  1. Phenotype file (contains non-genetic covariates if any) The file can be either space or tab-delimited with a header. It is required that the file contains one column for sample IDs and one column for the phenotype. It may contain columns for non-genetic covariates.

Note: Current version of SAIGE does not support categorical covariates that have more than two categories

less -S ./input/pheno_1000samples.txt_withdosages_withBothTraitTypes.txt
Output files
  1. model file
./output/example_quantitative.rda

#open R
R
#load the model file in R
load("./output/example_quantitative.rda")
names(modglmm)
modglmm$theta

#theta: a vector of length 2. The first element is the dispersion parameter estimate and the second one is the variance component parameter estimate
#coefficients: fixed effect parameter estimates
#linear.predictors: a vector of length N (N is the sample size) containing linear predictors
#fitted.values: a vector of length N (N is the sample size) containing fitted mean values on the original scale
#Y: a vector of length N (N is the sample size) containing final working vector
#residuals: a vector of length N (N is the sample size) containing residuals on the original scale
#sampleID: a vector of length N (N is the sample size) containing sample IDs used to fit the null model
  1. association result file for the subset of randomly selected markers
less -S ./output/example_quantitative_30markers.SAIGE.results.txt
  1. variance ratio file
less -S ./output/example_quantitative.varianceRatio.txt

Step 2: performing the single-variant association tests

  • For binary traits, saddle point approximation is used to account for case-control imbalance.
  • Four different file formats for dosages/genotypes of genetic variants to be tested can be used: VCF, BGEN, SAV, plain text
  • Conditional analysis based summary stats can be performed (--condition) can be performed in step 2 with dosage/genotype input file formats VCF, BGEN and SAV.
  • To test a subset of markers,
    • both variant IDs and range of chromosome positions can be specified for BGEN input (--idstoExcludeFile, --idstoIncludeFile, --rangestoExcludeFile, --rangestoIncludeFile)
    • range chromosome positions can be specified for VCF/SAV input (--chrom, --start, --end).
  • For VCF/SAV input, --chrom must be specified and the string needs to be exactly the same as in the VCF/SAV, such as "01" or "chr1".
  • For VCF/SAV input, --vcfField=DS to test dosages and --vcfField=GT to test genotypes
  • To drop samples with missing genotypes/dosages, --IsDropMissingDosages=TRUE, if FALSE, missing genotypes/dosages will be mean imputed. **NOTE: this option has not been extensively tested yet. **
#check the help info for step 2
Rscript step2_SPAtests.R --help

#perform the single-variant association tests
#for binary traits, --IsOutputAFinCaseCtrl=TRUE can be specified to output allele frequencies in cases and controls
Rscript step2_SPAtests.R	\
        --vcfFile=./input/genotype_10markers.missingness.vcf.gz \
        --vcfFileIndex=./input/genotype_10markers.missingness.vcf.gz.tbi \
        --vcfField=GT \
        --chrom=1 \
        --minMAF=0.0001 \
        --minMAC=1 \
        --sampleFile=./input/sampleIDindosage.txt \
        --GMMATmodelFile=./output/example_binary.rda \
        --varianceRatioFile=./output/example_binary.varianceRatio.txt \
        --SAIGEOutputFile=./output/example_binary.SAIGE.vcf.genotype.txt \
        --numLinesOutput=2 \
        --IsOutputAFinCaseCtrl=TRUE

##drop samples with missing genotypes/dosages
##--IsDropMissingDosages=TRUE
Rscript step2_SPAtests.R        \
        --vcfFile=./input/genotype_10markers.missingness.vcf.gz \
        --vcfFileIndex=./input/genotype_10markers.missingness.vcf.gz.tbi \
        --vcfField=GT \
        --chrom=1 \
        --minMAF=0.0001 \
        --minMAC=1 \
        --sampleFile=./input/sampleIDindosage.txt \
        --GMMATmodelFile=./output/example_binary.rda \
        --varianceRatioFile=./output/example_binary.varianceRatio.txt \
        --SAIGEOutputFile=./output/example_binary.SAIGE.vcf.genotype.dropmissing.txt \
        --numLinesOutput=2 \
        --IsOutputAFinCaseCtrl=TRUE	\
	--IsDropMissingDosages=TRUE    


#conditional analysis
## --condition = Genetic marker ids (chr:pos_ref/alt for vcf/sav or marker id for bgen) seperated by comma. e.g.chr3:101651171_C/T,chr3:101651186_G/A, Note that currently conditional analysis is only for vcf/sav and bgen input.
Rscript step2_SPAtests.R        \
        --vcfFile=./input/genotype_10markers.missingness.vcf.gz \
        --vcfFileIndex=./input/genotype_10markers.missingness.vcf.gz.tbi \
        --vcfField=GT \
        --chrom=1 \
        --minMAF=0.0001 \
        --minMAC=1 \
        --sampleFile=./input/sampleIDindosage.txt \
        --GMMATmodelFile=./output/example_binary.rda \
        --varianceRatioFile=./output/example_binary.varianceRatio.txt \
        --SAIGEOutputFile=./output/example_binary.SAIGE.vcf.genotype.dropmissing.txt \
        --numLinesOutput=2 \
        --IsOutputAFinCaseCtrl=TRUE     \
	--IsDropMissingDosages=TRUE	\
	--condition=1:4_A/C

#plain text input for dosages
Rscript step2_SPAtests.R \
        --dosageFile=./input/dosage_10markers.txt \
        --dosageFileNrowSkip=1 \
        --dosageFileNcolSkip=6 \
        --dosageFilecolnamesSkip=CHR,SNP,CM,POS,EFFECT_ALLELE,ALT_ALLELE \
        --minMAF=0.0001 \
        --sampleFile=./input/sampleIDindosage.txt \
        --GMMATmodelFile=./output/example.rda \
        --varianceRatioFile=./output/example.varianceRatio.txt \
        --SAIGEOutputFile=./output/example.plainDosage.SAIGE.txt \
        --numLinesOutput=2 \
        --IsOutputAFinCaseCtrl=TRUE

#bgen for dosages
Rscript step2_SPAtests.R \
        --bgenFile=./input/genotype_100markers.bgen \
        --bgenFileIndex=./input/genotype_100markers.bgen.bgi \
        --minMAF=0.0001 \
        --minMAC=1 \
        --sampleFile=./input/samplefileforbgen_10000samples.txt \
        --GMMATmodelFile=./output/example.rda \
        --varianceRatioFile=./output/example.varianceRatio.txt \
        --SAIGEOutputFile=./output/example.SAIGE.bgen.txt \
        --numLinesOutput=2 \
        --IsOutputAFinCaseCtrl=TRUE

#VCF containing genotypes
Rscript step2_SPAtests.R \
        --vcfFile=./input/genotype_10markers.vcf.gz \
        --vcfFileIndex=./input/genotype_10markers.vcf.gz.tbi \
        --vcfField=GT \
        --chrom=1 \
        --minMAF=0.0001 \
        --minMAC=1 \
        --sampleFile=./input/sampleIDindosage.txt \
        --GMMATmodelFile=./output/example.rda \
        --varianceRatioFile=./output/example.varianceRatio.txt \
        --SAIGEOutputFile=./output/example.SAIGE.vcf.genotype.txt \
        --numLinesOutput=2 \
        --IsOutputAFinCaseCtrl=TRUE

#VCF containing dosages
Rscript step2_SPAtests.R \
        --vcfFile=./input/dosage_10markers.vcf.gz \
        --vcfFileIndex=./input/dosage_10markers.vcf.gz.tbi \
        --vcfField=DS \
        --chrom=1 \
        --minMAF=0.0001 \
        --minMAC=1 \
        --sampleFile=./input/sampleIDindosage.txt \
        --GMMATmodelFile=./output/example.rda \
        --varianceRatioFile=./output/example.varianceRatio.txt \
        --SAIGEOutputFile=./output/example.SAIGE.vcf.dosage.txt \
        --numLinesOutput=2 \
        --IsOutputAFinCaseCtrl=TRUE

# SAV for dosages
Rscript step2_SPAtests.R        \
        --savFile=./input/dosage_10markers.sav  \
        --savFileIndex=./input/dosage_10markers.sav.s1r \
        --vcfField=DS \
        --minMAF=0.0001 \
        --minMAC=1 \
        --chrom=1 \
        --sampleFile=./input/samplefileforbgen_10000samples.txt \
        --GMMATmodelFile=./output/example.rda \
        --varianceRatioFile=./output/example.varianceRatio.txt \
        --SAIGEOutputFile=./output/example.SAIGE.sav.txt \
        --numLinesOutput=2 \
        --IsOutputAFinCaseCtrl=TRUE
Input files
  1. Dosage file SAIGE can take in several different formats for dosages: plain text (gzipped file is supported), VCF, BCF, BGEN, and SAV.
    The example dosage files contain 10 markers for 1,000 samples
  • plain text file
less -S dosage_10markers.txt
  • BGEN
genotype_100markers.bgen
genotype_100markers.bgen.bgi
  • VCF containing genotypes
zcat genotype_10markers.vcf.gz | less -S
genotype_10markers.vcf.gz.tbi
  • VCF containing dosages
zcat dosage_10markers.vcf.gz | less -S
dosage_10markers.vcf.gz.tbi
  • SAV
dosage_10markers.sav
dosage_10markers.sav.s1r
  1. Sample file <br> This file contains one column for sample IDs corresponding to the sample order in the dosage file. No header is included. The option was originally for BGEN file that does not contain sample information.
less -S sampleIDindosage.txt
  1. Model file from step 1
./output/example.rda
  1. Variance ratio file from step 1
./output/example.varianceRatio.txt
Output file

A file with association test results

less -S ./output/example.SAIGE.vcf.dosage.txt

NOTE:

  • If the input format of dosages is plain text. The first several columns for header are provided by the user.
  • For plain text input for dosage/genotypes, association results are for the allele whose dosages/genotypes are provides. For VCF, BCF, SAV, or BGEN input, association results are for Allele2.
  • For binary traits, the header of the output file: CHR POS SNPID Allele1 Allele2 AC_Allele2 AF_Allele2 imputationInfo N BETA SE Tstat p.value p.value.NA Is.SPA.converge varT varTstar AF.Cases AF.Controls
  • For quantitative traits, the header of the output file: CHR POS SNPID Allele1 Allele2 AC_Allele2 AF_Allele2 imputationInfo N BETA SE Tstat p.value varT varTstar 
CHR: chromosome
POS: genome position 
SNPID: variant ID
Allele1: allele 1
Allele2: allele 2
AC_Allele2: allele count of allele 2
AF_Allele2: allele frequency of allele 2
imputationInfo: imputation info. If not in dosage/genotype input file, will output 1
N: sample size
BETA: effect size of allele 2
SE: standard error of BETA
Tstat: score statistic of allele 2
p.value: p value (with SPA applied for binary traits)
p.value.NA: p value when SPA is not applied (only for binary traits)
Is.SPA.converge: whether SPA is converged or not (only for binary traits)
varT: estimated variance of score statistic with sample relatedness incorporated
varTstar: variance of score statistic without sample relatedness incorporated
AF.Cases: allele frequency of allele 2 in cases (only for binary traits and if --IsOutputAFinCaseCtrl=TRUE)
AF.Controls: allele frequency of allele 2 in controls (only for binary traits and if --IsOutputAFinCaseCtrl=TRUE)
Tstat_cond, p.value_cond, varT_cond, BETA_cond, SE_cond: summary stats for conditional analysis

Region- or gene-based association tests (SAIGE-GENE)

Step 0: creating a sparse GRM

Note: This step is only needed for region- and gene-based tests (SAIGE-GENE)
The sparse GRM only needs to be created once for each data set, e.g. a biobank, and can be used for all different phenotypes as long as all tested samples are included in it.

Rscript createSparseGRM.R	\
	--plinkFile=./input/nfam_100_nindep_0_step1_includeMoreRareVariants_poly \
	--nThreads=4  \
	--outputPrefix=./output/sparseGRM	\
	--numRandomMarkerforSparseKin=2000	\
	--relatednessCutoff=0.125
Input file

Genotype file for constructing the genetic relationship matrix
SAIGE takes the PLINK binary file for the genotypes and assumes the file prefix is the same one for .bed, .bim. and .fam

nfam_100_nindep_0_step1_includeMoreRareVariants_poly.bed
nfam_100_nindep_0_step1_includeMoreRareVariants_poly.bim
nfam_100_nindep_0_step1_includeMoreRareVariants_poly.fam
Output files
  1. a file storing the sparse GRM
sparseGRM_relatednessCutoff_0.125_1000_randomMarkersUsed.sparseGRM.mtx

The sparse GRM file can be opened using the readMM function in the R library Matrix

  1. a file storing IDs of the samples in the sparse GRM
sparseGRM_relatednessCutoff_0.125_1000_randomMarkersUsed.sparseGRM.mtx.sampleIDs.txt

Step 1: fitting the null logistic/linear mixed model

  • IsSparseKin=TRUE must be specified for SAIGE-GENE
  • step 1 model result from the single-variant assoc test can be re-used, except that for gene-based tests, variance ratios for multiple MAC categories and a sparse GRM need to be used (IsSparseKin=TRUE).
  • with IsSparseKin=TRUE, no sparseGRMFile and sparseGRMSampleIDFile are specified (step 0 has not been done beforehand), a sparse GRM will be created based on the relatednessCutoff in this step.
  • with IsSparseKin=TRUE, sparseGRMFile and sparseGRMSampleIDFile can be used to specify a pre-calcuated sparse GRM and the sample ids for the sparse GRM (output from step 0). Tested samples would be a subset of samples in the pre-calcuated GRM.
  • To activate the variance ratio estimation based multiple MAC categories, --isCateVarianceRatio=TRUE
  • cateVarRatioMinMACVecExclude and cateVarRatioMaxMACVecInclude are used to specify the MAC categories
#by default 
 --cateVarRatioMinMACVecExclude=0.5,1.5,2.5,3.5,4.5,5.5,10.5,20.5
 --cateVarRatioMaxMACVecInclude=1.5,2.5,3.5,4.5,5.5,10.5,20.5
corresponding to
0.5 < MAC <=  1.5
1.5 < MAC <=  2.5
2.5 < MAC <=  3.5
3.5 < MAC <=  4.5
4.5 < MAC <=  5.5
5.5 < MAC <=  10.5
10.5 < MAC <=  20.5
20.5 < MAC

Rscript step1_fitNULLGLMM.R     \
	--plinkFile=./input/nfam_100_nindep_0_step1_includeMoreRareVariants_poly \
        --phenoFile=./input/pheno_1000samples.txt_withdosages_withBothTraitTypes.txt \
        --phenoCol=y_quantitative \
        --covarColList=x1,x2 \
        --sampleIDColinphenoFile=IID \
        --traitType=quantitative       \
        --invNormalize=TRUE     \
        --outputPrefix=./output/example_quantitative \
	--outputPrefix_varRatio=./output/example_quantitative_cate	\
	--sparseGRMFile=./output/example_binary_cate.varianceRatio.txt.sparseGRM.mtx    \
        --sparseGRMSampleIDFile=./output/example_binary.varianceRatio.txt.sparseGRM.mtx.sample  \
        --nThreads=4 \
        --LOCO=FALSE	\
	--skipModelFitting=FALSE \
        --IsSparseKin=TRUE      \
        --isCateVarianceRatio=TRUE
Input files

  1. (same as step 1 for single-variant assoc tests and step 0)
    Genotype file for constructing the genetic relationship matrix in the plink format

  2. a file storing the sparse GRM (optional, output by step 0)

sparseGRM_relatednessCutoff_0.125_1000_randomMarkersUsed.sparseGRM.mtx

The sparse GRM file can be opened using the readMM function in the R library Matrix

  1. a file storing IDs of the samples in the sparse GRM (optional, output by step 0)
sparseGRM_relatednessCutoff_0.125_1000_randomMarkersUsed.sparseGRM.mtx.sampleIDs.txt
Output files
  • same as the step 1 output by SAIGE for single-variant association tests
  1. model file
  2. association result file for the subset of randomly selected markers
  3. variance ratio file
  • specific to SAIGE-GENE
  1. sparse Sigma file
sparseGRM_relatednessCutoff_0.125_1000_randomMarkersUsed.sparseSigma.mtx

NOTE: the file contains the sparse Sigma matrix, which is ** NOT ** the sparse GRM. The sparse Sigma matrix is computed based on the sparse GRM and the results of step 1.

Step 2: performing the region- or gene-based association tests

  • The command line is the same as the step 2 for single-variant assoc tests, except that a group file is specified (--groupFile)
  • Each line is for one gene/set of variants. The first element is for gene/set name.
  • The rest of the line is for variant ids included in this gene/set. For vcf/sav, the genetic marker ids are in the format chr:pos_ref/alt. For bgen, the genetic marker ids should match the ids in the bgen file. Each element in the line is separated by tab.
  • IsSingleVarinGroupTest=TRUE is to perform single-variant assoc tests as well for markers included in the gene-based tests
  • Same as the single-variant assoc tests, conditional analysis based summary stats can be performed (--condition) can be performed in step 2 with dosage/genotype input file formats VCF, BGEN and SAV.
Rscript step2_SPAtests.R \
        --vcfFile=./input/genotype_10markers.vcf.gz \
        --vcfFileIndex=./input/genotype_10markers.vcf.gz.tbi \
        --vcfField=GT \
        --chrom=1 \
	--minMAF=0 \
        --minMAC=0.5 \
        --maxMAFforGroupTest=0.01       \
        --sampleFile=./input/samplelist.txt \
        --GMMATmodelFile=./output/example_quantitative.rda \
        --varianceRatioFile=./output/example_quantitative_cate.varianceRatio.txt \
        --SAIGEOutputFile=./output/example_quantitative.SAIGE.gene.txt \
        --numLinesOutput=1 \
        --groupFile=./input/groupFile_geneBasedtest_simulation.txt    \
        --sparseSigmaFile=./output/example_quantitative_cate.varianceRatio.txt_relatednessCutoff_0.125.sparseSigma.mtx       \
        --IsSingleVarinGroupTest=TRUE


##another example, conditional analysis for gene-based tests
Rscript step2_SPAtests.R \
        --vcfFile=./input/seedNumLow_126001_seedNumHigh_127000_nfam_1000_nindep_0.sav \
        --vcfFileIndex=./input/seedNumLow_126001_seedNumHigh_127000_nfam_1000_nindep_0.sav.s1r \
        --vcfField=DS \
        --chrom=chr1 \
        --minMAF=0 \
        --minMAC=0.5 \
        --maxMAFforGroupTest=0.01       \
        --sampleFile=./input/samplelist.txt \
        --GMMATmodelFile=./output/example_quantitative.rda \
        --varianceRatioFile=./output/example_quantitative_cate.varianceRatio.txt \
	--SAIGEOutputFile=./output/example_quantitative.SAIGE.gene_conditional.txt \
        --numLinesOutput=1 \
        --groupFile=./input/groupFile_geneBasedtest.txt    \
        --sparseSigmaFile=./output/example_quantitative_cate.varianceRatio.txt_relatednessCutoff_0.125.sparseSigma.mtx       \
        --IsOutputAFinCaseCtrl=TRUE     \
        --IsSingleVarinGroupTest=TRUE   \
	--condition=chr1:32302_A/C
Input files

  1. Dosage file
    SAIGE-GENE can take in VCF, BCF, BGEN, and SAV. The plain text dosage/genotype input does not work.

  2. Sample file
    This file contains one column for sample IDs corresponding to the sample order in the dosage file. No header is included.

less -S sampleIDindosage.txt
  1. Model file from step 1 
./output/example.rda
  1. Variance ratio file from step 1 
./output/example.varianceRatio.txt
  • specific to SAIGE-GENE
  1. Group file
    Each line is for one gene/set of variants. The first element is for gene/set name.
    The rest of the line is for variant ids included in this gene/set. For vcf/sav, the genetic marker ids are in the format chr:pos_ref/alt. For bgen, the genetic marker ids should match the ids in the bgen file. Each element in the line is separated by tab.
less -S ./input/groupFile_geneBasedtest.txt
Output files
  1. A file with region- or gene-based association test results
less -S ./output/example_quantitative.SAIGE.gene_conditional.txt

NOTE:

  • The header of the output file: Gene Pvalue Pvalue_cond Nmarker_MACCate_1 Nmarker_MACCate_2 Nmarker_MACCate_3 Nmarker_MACCate_4 Nmarker_MACCate_5 Nmarker_MACCate_6 Nmarker_MACCate_7 Nmarker_MACCate_8 markerIDs markerAFs Pvalue_Burden Pvalue_SKAT Pvalue_Burden_cond Pvalue_SKAT_cond 
Gene : Gene name (as provided in the group file)
Pvalue: p-value of SKAT-O test
Pvalue_cond: conditional p-value of SKAT-O test (if --condition= is specified)
Nmarker_MACCate_n: number of markers in the gene falling in the nth MAC category (The MAC category is corresponding to the one used for the variance ratio estimation). For example, by default, Nmarker_MACCate_1 is the number of singletons
markerIDs: IDs for variants included in the test
markerAFs: allele frequencies for variants included in the test
Pvalue_Burden: p-value of Burden test
Pvalue_SKAT: p-value of SKAT test
Pvalue_Burden_cond: conditional p-value of Burden test (if --condition= is specified)
Pvalue_SKAT_cond: conditional p-value of SKAT test (if --condition= is specified)
  1. A file with single-variant association test results for genetic variants included in the gene-based tests (if --IsSingleVarinGroupTest=TRUE)
less -S ./output/example_quantitative.SAIGE.gene_conditional.txt_single

Same as the output by SAIGE for single-variant association tests

Clone this wiki locally