BayesC
Last updated: 2023-09-26
Checks: 6 1
Knit directory: dgrp-starve/
This reproducible R Markdown analysis was created with workflowr (version 1.7.1). The Checks tab describes the reproducibility checks that were applied when the results were created. The Past versions tab lists the development history.
Great! Since the R Markdown file has been committed to the Git repository, you know the exact version of the code that produced these results.
Great job! The global environment was empty. Objects defined in the global environment can affect the analysis in your R Markdown file in unknown ways. For reproduciblity it’s best to always run the code in an empty environment.
The command set.seed(20221101) was run prior to running the code in the R Markdown file. Setting a seed ensures that any results that rely on randomness, e.g. subsampling or permutations, are reproducible.
Great job! Recording the operating system, R version, and package versions is critical for reproducibility.
Nice! There were no cached chunks for this analysis, so you can be confident that you successfully produced the results during this run.
Using absolute paths to the files within your workflowr project makes it difficult for you and others to run your code on a different machine. Change the absolute path(s) below to the suggested relative path(s) to make your code more reproducible.
| absolute | relative |
|---|---|
| /data2/morgante_lab/nklimko/rep/dgrp-starve/ | . |
Great! You are using Git for version control. Tracking code development and connecting the code version to the results is critical for reproducibility.
The results in this page were generated with repository version 82f6ff8. See the Past versions tab to see a history of the changes made to the R Markdown and HTML files.
Note that you need to be careful to ensure that all relevant files for the analysis have been committed to Git prior to generating the results (you can use wflow_publish or wflow_git_commit). workflowr only checks the R Markdown file, but you know if there are other scripts or data files that it depends on. Below is the status of the Git repository when the results were generated:
Ignored files:
Ignored: .snakemake/
Ignored: code/methodComp/bglr/err-bglr-f.5381.err
Ignored: code/methodComp/bglr/err-bglr-m.5382.err
Ignored: code/methodComp/m/meth-m.4685.err
Ignored: data/fb/
Ignored: data/snake/
Ignored: snake/.snakemake/
Ignored: snake/ReadMe.md
Ignored: snake/Snakefile.yaml
Ignored: snake/code/misc/
Ignored: snake/data/
Ignored: snake/dgrp.yaml
Ignored: snake/logs/
Ignored: snake/slurm/
Ignored: snake/snakemake_submitter.sbatch
Ignored: snake/zz_lost/
Ignored: zz_lost/
Untracked files:
Untracked: code/methodComp/m/meth-m.4676.err
Untracked: code/methodComp/method-f.4751.out
Unstaged changes:
Modified: .Rprofile
Modified: .gitattributes
Modified: analysis/Method/BayesC.Rmd
Note that any generated files, e.g. HTML, png, CSS, etc., are not included in this status report because it is ok for generated content to have uncommitted changes.
These are the previous versions of the repository in which changes were made to the R Markdown (analysis/man-BayesC.Rmd) and HTML (docs/man-BayesC.html) files. If you’ve configured a remote Git repository (see ?wflow_git_remote), click on the hyperlinks in the table below to view the files as they were in that past version.
| File | Version | Author | Date | Message |
|---|---|---|---|---|
| html | 82f6ff8 | nklimko | 2023-09-26 | Build site. |
| Rmd | bc93211 | nklimko | 2023-09-26 | comprehensiv e commit after 100mb fix |
| html | bc93211 | nklimko | 2023-09-26 | comprehensiv e commit after 100mb fix |
#reorders 50 X m table to a (50*m) X 2 table
#converts method into factor to retain order
ggTidy <- function(data){
for(i in 1:dim(data)[2]){
name <- colnames(data)[i]
temp <- cbind(rep(name, dim(data)[1]), data[,i, with=FALSE])
if(i==1){
hold <- temp
} else{
hold <- rbind(hold, temp, use.names=FALSE)
}
}
colnames(hold) <- c("method", "cor")
hold$method <- factor(hold$method, levels=unique(hold$method))
return(hold)
}
#wrapper for ggplot call to custom fill sex, title, and y axis label
ggMake <- function(data, sex, custom.title, custom.Ylab){
plothole <- ggplot(data, aes(x=method, y=cor, fill=method)) +
geom_violin(color = NA, width = 0.65) +
geom_boxplot(color='#440154FF', width = 0.15) +
theme_minimal() +
stat_summary(fun=mean, color='#440154FF', geom='point',
shape=18, size=3, show.legend=FALSE) +
labs(x=NULL,y=custom.Ylab, tag=sex, title=custom.title) +
theme(legend.position='none',
axis.text.x = element_text(angle = -45, size=10),
text=element_text(size=10),
plot.tag = element_text(size=15)) +
scale_fill_viridis(begin = 0.4, end=0.9,discrete=TRUE)
return(plothole)
}#bayesFinal <- readRDS("snake/data/zigure/bayesFinal.Rds")
b1 <- readRDS("snake/data/sr/34_opcor/f/bayesC_0.1.Rds")
b5 <- readRDS("snake/data/sr/34_opcor/f/bayesC_0.5.Rds")
b9 <- readRDS("snake/data/sr/34_opcor/f/bayesC_0.9.Rds")
bayesOp <- data.table(b1, b5, b9)
colnames(bayesOp) <- c('0.1', '0.5', '0.9')
data <- na.omit(ggTidy(bayesOp))
gg[[1]] <- ggMake(data, 'F', 'BayesC with Variable R2', 'Correlation')b1 <- readRDS("snake/data/sr/34_opcor/m/bayesC_0.1.Rds")
b5 <- readRDS("snake/data/sr/34_opcor/m/bayesC_0.5.Rds")
b9 <- readRDS("snake/data/sr/34_opcor/m/bayesC_0.9.Rds")
bayesOp <- data.table(b1, b5, b9)
colnames(bayesOp) <- c('0.1', '0.5', '0.9')
data <- na.omit(ggTidy(bayesOp))
gg[[2]] <- ggMake(data, 'M', 'BayesC with Variable R2', 'Correlation')base <- 'data/bglr/f/trace_'
base <- 'snake/data/bglr/f/trace_'
ETA <- paste0(base, 'ETA_1_parBayesC.dat')
varPath <- paste0(base, 'varE.dat')
fitPath <- paste0(base, 'fitBGLR')
fit <- readRDS(fitPath)
# Residual Variance
varE<-scan(varPath)
varFrame <- as.data.frame(varE)
colnames(varFrame) <- 'varVal'
gg[[3]] <- ggplot(varFrame, aes(x = 1:nrow(varFrame), y = varVal)) +
labs(x=NULL,y='varE', tag='F', title='Residual Variance') +
geom_point() +
geom_hline(yintercept = fit$varE, color='blue') +
geom_vline(xintercept = fit$burnIn/fit$thin, color='blue')
# Samples
TMP=read.table(ETA, header=T)
tmpFrame <- as.data.frame(TMP)
colnames(tmpFrame) <- 'tmp'
gg[[5]] <- ggplot(tmpFrame, aes(x = 1:nrow(tmpFrame), y = tmp)) +
labs(x=NULL,y='TMP', tag='F', title='Samples') +
geom_point() +
geom_hline(yintercept = fit$ETA[[1]]$probIn, color='blue') +
geom_vline(xintercept = fit$burnIn/fit$thin, color='blue')base <- 'data/bglr/m/trace_'
base <- 'snake/data/bglr/m/trace_'
ETA <- paste0(base, 'ETA_1_parBayesC.dat')
varPath <- paste0(base, 'varE.dat')
fitPath <- paste0(base, 'fitBGLR')
fit <- readRDS(fitPath)
# Residual Variance
varE<-scan(varPath)
varFrame <- as.data.frame(varE)
colnames(varFrame) <- 'varVal'
gg[[4]] <- ggplot(varFrame, aes(x = 1:nrow(varFrame), y = varVal)) +
labs(x=NULL,y='varE', tag='M', title='Residual Variance') +
geom_point() +
geom_hline(yintercept = fit$varE, color='blue') +
geom_vline(xintercept = fit$burnIn/fit$thin, color='blue')
# Samples
TMP=read.table(ETA, header=T)
tmpFrame <- as.data.frame(TMP)
colnames(tmpFrame) <- 'tmp'
gg[[6]] <- ggplot(tmpFrame, aes(x = 1:nrow(tmpFrame), y = tmp)) +
labs(x=NULL,y='TMP', tag='M', title='Samples') +
geom_point() +
geom_hline(yintercept = fit$ETA[[1]]$probIn, color='blue') +
geom_vline(xintercept = fit$burnIn/fit$thin, color='blue')
# Residual variance
#varE<-scan(varPath)
#plot(varE,type='o',col=2,cex=.5,ylab=expression(var[e]));
#abline(h=fit$varE,col=4,lwd=2)
#abline(v=fit$burnIn/fit$thin,col=4)
# Samples
#TMP=read.table(ETA, header=T)
#plot(TMP[,1],type='o',col=4, cex=0.5)
#abline(h=fit$ETA[[1]]$probIn,col=4,lwd=2)
#abline(v=fit$burnIn/fit$thin,col=4)Intro to Method
- BayesC - BayesC uses a spike-and-slab prior to perform effect shrinkage and variable selection. The model is fit using Markov Chain Monte Carlo(MCMC) methods.
Parameters
For BayesC, we defined the following parameters: - nIter : number of sampling iterations - burnIn : iteration burn number, discarded from front end of sampling - thin : factor to scale down nIter and burnIn for memory conservation - R2 : proportion of variance explained by the model, 0 < R2 < 1
nIter, burnIn, and thin were set to 130k, 30k, and 50 respectively. These parameters affect Markov Chain Monte Carlo sampling to generate a distribution of effect weights.
R2 was chosen to be 0.9 after testing low, intermediate, and high values for R2.
Results
Proof of R2
plot_grid(gg[[1]], gg[[2]], ncol=2)
Proof of niter/burnin/thin
The following trace plots show the residual variance and samples for a single run:
Male
plot_grid(gg[[3]], gg[[5]], ncol=2)
Female
plot_grid(gg[[4]], gg[[6]], ncol=2)
sessionInfo()R version 4.1.2 (2021-11-01)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Rocky Linux 8.5 (Green Obsidian)
Matrix products: default
BLAS/LAPACK: /opt/ohpc/pub/libs/gnu9/openblas/0.3.7/lib/libopenblasp-r0.3.7.so
locale:
[1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
[3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
[5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
[7] LC_PAPER=en_US.UTF-8 LC_NAME=C
[9] LC_ADDRESS=C LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] reshape2_1.4.4 melt_1.10.0 ggcorrplot_0.1.4.1 lubridate_1.9.2
[5] forcats_1.0.0 stringr_1.5.0 purrr_1.0.1 readr_2.1.4
[9] tidyr_1.3.0 tibble_3.2.1 tidyverse_2.0.0 scales_1.2.1
[13] viridis_0.6.4 viridisLite_0.4.2 qqman_0.1.9 cowplot_1.1.1
[17] ggplot2_3.4.3 data.table_1.14.8 dplyr_1.1.3 workflowr_1.7.1
loaded via a namespace (and not attached):
[1] Rcpp_1.0.11 getPass_0.2-2 ps_1.7.5 rprojroot_2.0.3
[5] digest_0.6.33 utf8_1.2.3 plyr_1.8.8 R6_2.5.1
[9] evaluate_0.21 httr_1.4.7 highr_0.10 pillar_1.9.0
[13] rlang_1.1.1 rstudioapi_0.15.0 whisker_0.4.1 callr_3.7.3
[17] jquerylib_0.1.4 rmarkdown_2.23 labeling_0.4.3 munsell_0.5.0
[21] compiler_4.1.2 httpuv_1.6.11 xfun_0.39 pkgconfig_2.0.3
[25] htmltools_0.5.5 tidyselect_1.2.0 gridExtra_2.3 fansi_1.0.4
[29] calibrate_1.7.7 tzdb_0.4.0 withr_2.5.0 later_1.3.1
[33] MASS_7.3-60 grid_4.1.2 jsonlite_1.8.7 gtable_0.3.4
[37] lifecycle_1.0.3 git2r_0.32.0 magrittr_2.0.3 cli_3.6.1
[41] stringi_1.7.12 cachem_1.0.8 farver_2.1.1 fs_1.6.3
[45] promises_1.2.0.1 bslib_0.5.0 generics_0.1.3 vctrs_0.6.3
[49] tools_4.1.2 glue_1.6.2 hms_1.1.3 processx_3.8.2
[53] fastmap_1.1.1 yaml_2.3.7 timechange_0.2.0 colorspace_2.1-0
[57] knitr_1.43 sass_0.4.7