improve example

example/README.md

@@ -0,0 +1,53 @@
+README
+======
+
+Here we present several exemplar command lines to use rvtests.
+A complete manual can be found: 
+[https://zhanxw.github.io/rvtests/](https://zhanxw.github.io/rvtests/)
+
+= run single variant test perform wald test
+
+    ../executable/rvtest --pheno pheno --inVcf example.vcf --single wald --out out1
+
+
+= using the second column of phenotype
+
+    ../executable/rvtest --pheno pheno --inVcf example.vcf --single wald --mpheno 2 --out out2
+
+
+= using the phenotype y2
+
+    ../executable/rvtest --pheno pheno --inVcf example.vcf --single wald --pheno-name y2 --out out3
+
+
+= use covariate c1,c2
+
+    ../executable/rvtest --pheno pheno --inVcf example.vcf --single wald --covar covar --covar-name c1,c2 --out out4
+
+= use covariate file with missing data
+
+    ../executable/rvtest --pheno pheno --inVcf example.vcf --single wald --covar covar.missing --covar-name c1,c2 --out out5
+
+= produce meta-analysis statistics for unrelated individuals, using covariates, regress covariates on pheontpye, then inverse normalize residuals, perform score test and generate covariance matrices
+
+    ../executable/rvtest --pheno pheno --inVcf example.vcf --meta score,cov --covar covar --covar-name c1,c2 --useResidualAsPhenotype --inverseNormal --out out6
+
+= generate kinship matrix from pedigree
+
+    ../executable/vcf2kinship --ped pheno --bn --out output
+
+= generate empirical kinship matrix from VCF file using Balding-Nicols method
+
+    ../executable/vcf2kinship --inVcf example.vcf --bn --out output
+
+= produce meta-analysis statistics for related individuals, using covariates, regress covariates on pheontpye, then inverse normalize residuals, perform score test and generate covariance matrices
+
+need to use a kinship matrix (see previous steps) output.kinship
+then, use:
+
+    ../executable/rvtest --pheno pheno --inVcf example.vcf --meta score,cov --covar covar --covar-name c1,c2 --useResidualAsPhenotype --inverseNormal --kinship output.kinship --out out7
+
+
+= produce meta-analysis statistics under dominant and recessive model for unrelated individuals, using covariates (c1, c2 columns in covar) and phenotype (y4 column in pheno)
+
+    ../executable/rvtest --pheno pheno --inVcf example.vcf --meta score,cov,dominant,recessive --covar covar --covar-name c1,c2 --out out8 --pheno-name y4

example/cmd.sh

@@ -1,5 +1,11 @@
+#!/bin/bash
 # each command here is an example of how to use rvtests
-# see README for more information
+# see README.md for more information
+
+if [ ! -e ../executable/rvtest ]; then
+    (cd ../; make)
+fi
+
 ../executable/rvtest --pheno pheno --inVcf example.vcf --single wald --out out1
 ../executable/rvtest --pheno pheno --inVcf example.vcf --single wald --mpheno 2 --out out2
 ../executable/rvtest --pheno pheno --inVcf example.vcf --single wald --pheno-name y2 --out out3