Add glomeruli data download from greenelab/rheum-plier-data (#31)

* Add glom data download from data repo

Ignore downloaded files

* Update: rerun DELV with data downloaded from data repo

* Update: rerun with data from data repo

* Update: rerun figure notebook with data downloaded from data repo

No changes in p-values on display items

.gitignore

@@ -33,12 +33,14 @@ data/expression_data/E-MTAB-2452_hugene11st_SCANfast.pcl
 data/expression_data/GSE18885_series_matrix.txt
 data/expression_data/NARES_SCANfast_ComBat.pcl
 data/expression_data/SLE_WB_all_microarray_QN_zto_before.pcl
+data/expression_data/ERCB_Glom_CustCDF19_forVCRC.txt
 data/sample_info/E-GEOD-39088.sdrf.txt
 data/sample_info/E-GEOD-65391.sdrf.txt
 data/sample_info/E-GEOD-78193.sdrf.txt
 data/sample_info/E-MTAB-2452.sdrf.txt
 data/sample_info/NARES_demographic_data.tsv
 data/sample_info/sle-wb_sample_dataset_mapping.tsv
+data/sample_info/ERCB_glom_diagnosis.tsv
 
 # private data
 data/kidney/

00-data_download.sh

@@ -27,6 +27,9 @@ wget https://github.com/greenelab/rheum-plier-data/raw/4be547553f24fecac9e2f5c2b
 # isolated blood cell populations from autoimmune conditions
 wget https://github.com/greenelab/rheum-plier-data/raw/4be547553f24fecac9e2f5c2b469a17f9df253f0/isolated-cell-pop/processed/E-MTAB-2452_hugene11st_SCANfast.pcl
 
+# glomeruli data
+wget https://github.com/greenelab/rheum-plier-data/raw/55d86bb537f9e38c83fc3cca993cde48dc984411/glomeruli/ERCB_Glom_CustCDF19_forVCRC.txt
+
 # get sample (e.g., phenotype) data
 cd .. && mkdir sample_info && cd sample_info
 # sle-wb sample to dataset of origin data
@@ -37,3 +40,5 @@ wget https://github.com/greenelab/rheum-plier-data/raw/4be547553f24fecac9e2f5c2b
 wget https://github.com/greenelab/rheum-plier-data/raw/4be547553f24fecac9e2f5c2b469a17f9df253f0/sle-wb/arrayexpress/E-GEOD-39088/E-GEOD-39088.sdrf.txt
 wget https://github.com/greenelab/rheum-plier-data/raw/4be547553f24fecac9e2f5c2b469a17f9df253f0/sle-wb/arrayexpress/E-GEOD-78193/E-GEOD-78193.sdrf.txt
 wget https://github.com/greenelab/rheum-plier-data/raw/4be547553f24fecac9e2f5c2b469a17f9df253f0/NARES/NARES_demographic_data.tsv
+wget https://github.com/greenelab/rheum-plier-data/raw/55d86bb537f9e38c83fc3cca993cde48dc984411/glomeruli/ERCB_glom_diagnosis.tsv
+

20-kidney_differential_expression.Rmd

@@ -35,7 +35,7 @@ dir.create(results.dir, recursive = TRUE, showWarnings = FALSE)
 ### Expression data
 
 ```{r}
-ercb.file <- file.path("data", "kidney", 
+ercb.file <- file.path("data", "expression_data", 
                        "ERCB_Glom_CustCDF19_forVCRC.txt")
 exprs.df <- readr::read_tsv(ercb.file)
 exprs.df <- dplyr::select(exprs.df, -EntrezGeneID)
@@ -46,7 +46,7 @@ colnames(exprs.df)[1] <- "Gene"
 agg.exprs.df <- PrepExpressionDF(exprs.df)
 # any genes that don't have a gene symbol need to be dropped
 agg.exprs.df <- dplyr::filter(agg.exprs.df, !(is.na(Gene)))
-readr::write_tsv(agg.exprs.df, path = file.path("data", "kidney", 
+readr::write_tsv(agg.exprs.df, path = file.path("data", "expression_data", 
                                                 "ERCB_Glom_mean_agg.pcl"))
 # as a matrix for use with PLIER
 exprs.mat <- as.matrix(dplyr::select(agg.exprs.df, -Gene))
@@ -56,36 +56,8 @@ rownames(exprs.mat) <- agg.exprs.df$Gene
 ### Clinical data
 
 ```{r}
-clinical.file <- file.path("data", "kidney", 
-                           "Neptune_ERCB_GE_Clinical_Data_2016-08-24_15-11-58-1.txt")
-clinical.df <- readr::read_tsv(clinical.file)
-```
-
-```{r}
-# only retain samples that are in the expression data
-microarray.samples <- colnames(exprs.mat)
-all(microarray.samples %in% clinical.df$`Microarray Sample ID`)
-```
-
-```{r}
-# we only want the info for the samples in the ERCB glomeruli data -- that's
-# the microarray data we're looking at
-clinical.df <- clinical.df %>% 
-                 dplyr::filter(`Microarray Sample ID` %in% microarray.samples,
-                               `Tissue Source` == "Glomeruli")
-
-# diagnosis information only - this will serve as group labels
-# recode diagnosis such that nephrotic syndrome diagnoses are grouped
-diagnosis.df <- clinical.df %>%
-  dplyr::select(c(`Microarray Sample ID`, Diagnosis)) %>%
-  dplyr::mutate(Diagnosis = 
-                  dplyr::recode(Diagnosis,
-                                MCD = "Nephrotic syndrome",
-                                MN = "Nephrotic syndrome",
-                                FSGS = "Nephrotic syndrome",
-                                `FSGS/MCD` = "Nephrotic syndrome"),
-                Sample = `Microarray Sample ID`) %>%
-  dplyr::select(c(Sample, Diagnosis))
+clinical.file <- file.path("data", "sample_info", "ERCB_glom_diagnosis.tsv")
+diagnosis.df <- readr::read_tsv(clinical.file)
 ```
 
 ## Apply recount2 model
@@ -101,7 +73,7 @@ recount.b <- GetNewDataB(exprs.mat = as.matrix(exprs.mat),
 
 ```{r}
 # save B matrix to file
-b.file <- file.path("data", "kidney", "ERCB_glom_recount2_B.RDS")
+b.file <- file.path(results.dir, "ERCB_glom_recount2_B.RDS")
 saveRDS(recount.b, file = b.file)
 ```
 

21-AAV_DLVE.Rmd

@@ -19,14 +19,6 @@ At first, we'll take the naive approach of just looking at the overlapping
 sets of significant LVs.
 There's no guarantee that the directionality will be in agreement this way.
 
-## Install VennDiagram & cowplot
-
-```{r}
-# current versions on CRAN
-devtools::install_url("https://cran.r-project.org/src/contrib/VennDiagram_1.6.20.tar.gz")
-devtools::install_url("https://cran.r-project.org/src/contrib/cowplot_0.9.2.tar.gz")
-```
-
 ## Functions and directory set up
 
 ```{r}

plots/21/AAV_FDR_0.05_overlap.png.2018-08-15_20-45-43.log

@@ -0,0 +1,15 @@
+INFO [2018-08-15 20:45:43] $x
+INFO [2018-08-15 20:45:43] sig.list
+INFO [2018-08-15 20:45:43] 
+INFO [2018-08-15 20:45:43] $filename
+INFO [2018-08-15 20:45:43] vd.file
+INFO [2018-08-15 20:45:43] 
+INFO [2018-08-15 20:45:43] $imagetype
+INFO [2018-08-15 20:45:43] [1] "png"
+INFO [2018-08-15 20:45:43] 
+INFO [2018-08-15 20:45:43] $resolution
+INFO [2018-08-15 20:45:43] [1] 600
+INFO [2018-08-15 20:45:43] 
+INFO [2018-08-15 20:45:43] $category.names
+INFO [2018-08-15 20:45:43] c("NARES", "GPA blood", "ERCB glomeruli")
+INFO [2018-08-15 20:45:43]