Add files via upload

For_MASH/mashr_eQTL.R

@@ -0,0 +1,129 @@
+library(ashr)
+library(mashr)
+library(corrplot)
+
+
+dat <- readRDS("formatted/FastQTLSumStats.list.mash.rds")
+names(dat)
+dim(dat$strong.b)
+dim(dat$strong.z)
+dim(dat$strong.s)
+
+dim(dat$random.b)
+dim(dat$random.z)
+dim(dat$random.s)
+
+
+data.strong = list(Bhat=dat$strong.b,  Shat=dat$strong.s )
+data.random = list(Bhat=dat$random.b,  Shat=dat$random.s )
+
+m.1by1 = mash_1by1(mash_set_data(data.strong$Bhat, data.strong$Shat))
+strong.subset = get_significant_results(m.1by1,0.05)
+
+random.subset = sample(1:nrow(data.random$Bhat), 10000)
+
+data.temp = mash_set_data(data.random$Bhat[random.subset,],  data.random$Shat[random.subset,])
+Vhat = estimate_null_correlation_simple(data.temp)
+rm(data.temp)
+
+
+
+
+data.random2 = mash_set_data(data.random$Bhat[random.subset,],  data.random$Shat[random.subset,], V=Vhat)
+data.strong2 = mash_set_data(data.strong$Bhat[strong.subset,],  data.strong$Shat[strong.subset,], V=Vhat)
+
+
+U.pca = cov_pca(data.strong2, 5)
+U.ed = cov_ed(data.strong2, U.pca)
+
+
+U.c = cov_canonical(data.random2)
+m = mash(data.random2, Ulist = c(U.ed,U.c), outputlevel = 1)
+
+m2 = mash(data.strong2, g=get_fitted_g(m), fixg=TRUE)
+
+head(get_lfsr(m2))
+head(get_pm(m2))
+head(get_psd(m2))
+dim(get_lfsr(m2))
+dim(get_pm(m2))
+dim(get_psd(m2))
+
+
+
+results = get_significant_results(m2, thresh = 0.001, conditions = NULL, sig_fn = get_lfsr )
+
+print(  length(results) ) 
+
+pdf("top5.pdf")
+mash_plot_meta(m.c,get_significant_results(m.c)[1], ylab = "Brain regions")
+mash_plot_meta(m.c,get_significant_results(m.c)[2], ylab = "Brain regions")
+mash_plot_meta(m.c,get_significant_results(m.c)[3], ylab = "Brain regions")
+mash_plot_meta(m.c,get_significant_results(m.c)[4], ylab = "Brain regions")
+mash_plot_meta(m.c,get_significant_results(m.c)[5], ylab = "Brain regions")
+dev.off()
+
+
+
+m3 = mash_compute_posterior_matrices(m2, data.strong2, algorithm.version = 'R',  posterior_samples = 100)
+names(m3)
+m2$result = m3
+x3 = get_pairwise_sharing(m2, factor=0.5)
+
+
+pdf("Pairwise_sharing.pdf")
+corrplot(x3, method='color', cl.lim=c(0,1), type='upper', addCoef.col = "black", tl.col="black", tl.srt=45, title = 'Pairwise Sharing by Magnitude', mar = c(4,0,4,0))
+dev.off()
+
+
+
+x3a = get_pairwise_sharing(m2, factor=0.5, lfsr_thresh = 1)
+pdf("Pairwise_sharing-2.pdf")
+corrplot(x3a, method='color', cl.lim=c(0,1), type='upper', addCoef.col = "black", tl.col="black", tl.srt=45, title = 'Pairwise Sharing by Magnitude', mar = c(4,0,4,0))
+dev.off()
+
+
+
+x3b = get_pairwise_sharing(m2, factor=0.5, lfsr_thresh = 0.05)
+pdf("Pairwise_sharing-3.pdf")
+corrplot(x3b, method='color', cl.lim=c(0,1), type='upper', addCoef.col = "black", tl.col="black", tl.srt=45, title = 'Pairwise Sharing by Magnitude', mar = c(4,0,4,0))
+dev.off()
+
+
+
+
+x3a = get_pairwise_sharing_from_samples(m2, factor=0.5, lfsr_thresh = 1)
+pdf("Pairwise_sharing-4.pdf")
+corrplot(x3a, method='color', cl.lim=c(0,1), type='upper', addCoef.col = "black", tl.col="black", tl.srt=45, title = 'Pairwise Sharing by Magnitude', mar = c(4,0,4,0))
+dev.off()
+
+
+
+
+x3b = get_pairwise_sharing_from_samples(m2, factor=0.5, lfsr_thresh = 0.05)
+pdf("Pairwise_sharing-5.pdf")
+corrplot(x3b, method='color', cl.lim=c(0,1), type='upper', addCoef.col = "black", tl.col="black", tl.srt=45, title = 'Pairwise Sharing by Magnitude', mar = c(4,0,4,0))
+dev.off()
+
+
+
+
+
+
+
+
+
+
+
+x3b = get_pairwise_sharing(m2, factor=0.5, lfsr_thresh = 0.001)
+pdf("Pairwise_sharing-6.pdf")
+corrplot(x3b, method='color', cl.lim=c(0,1), type='upper', addCoef.col = "black", tl.col="black", tl.srt=45, title = 'Pairwise Sharing by Magnitude', mar = c(4,0,4,0))
+dev.off()
+
+
+
+
+
+
+
+