1 Setup

CRC paper actually used both training and validation datasets (total of 18) for Figure 4. ~~Fig4C_CMSvs.R and Fig4C_contScores.R scripts are using only 10 validation datasets. If you want to use all 18 datasets, block the below lines in those scripts.~~ The current version of vignette uses all 18 studies.

# load("data/eSets/trainingSetNames.RData")
# validationSetNames <- setdiff(setNames, trainingSetNames)
# setNames <- validationSetNames

Fig4C_CMSvs.R script requires three environmental variables set before running: m2_name (column name for the model to be compared to CMS), m2_1 and m2_2 (names of the two model variables that form the group to be compared to CMS)
Fig4C_contScores.R script requires six environmental variables set before running: m1_name, m1_1, m1_2 for one model, m2_name, m2_1, m2_2 for the other.

2 Compare to CMS

2.1 CMS vs. PCSS1/2

m2_name <- "PCSS"
m2_1 <- "PCSS1"
m2_2 <- "PCSS2"
source("R/Fig4C_CMSvs.R", print.eval = TRUE)
## Warning: Removed 86 rows containing non-finite values (stat_boxplot).
## Warning: Removed 86 rows containing missing values (geom_point).

2.2 CMS vs. RAV1575/834

RAV1575/834 are the most similar PCclsuters to PCSS1/2, respectively, based on Pearson correlation.

m2_name <- "RAV"
m2_1 <- "RAV1575"
m2_2 <- "RAV834"
source("R/Fig4C_CMSvs.R", print.eval = TRUE)
## Warning: Removed 86 rows containing non-finite values (stat_boxplot).
## Warning: Removed 86 rows containing missing values (geom_point).

2.3 CMS vs. RAV834/833

RAV834/833 have the largest r-squared score when we compared the samples scores against the metadata, CMS.

m2_name <- "RAV"
m2_1 <- "RAV834"
m2_2 <- "RAV833"
source("R/Fig4C_CMSvs.R", print.eval = TRUE)
## Warning: Removed 86 rows containing non-finite values (stat_boxplot).
## Warning: Removed 86 rows containing missing values (geom_point).

3 Compare continuous scores

3.1 PCSS vs. RAV1575/834

m1_name <- "PCSS"
m1_1 <- "PCSS1"
m1_2 <- "PCSS2"
m2_name <- "RAV"
m2_1 <- "RAV1575"
m2_2 <- "RAV834"

source("R/Fig4C_contScores.R", print.eval = TRUE)
## Warning: Removed 86 rows containing non-finite values (stat_boxplot).
## Warning: Removed 86 rows containing missing values (geom_point).

3.2 PCSS vs. RAV834/833

m1_name <- "PCSS"
m1_1 <- "PCSS1"
m1_2 <- "PCSS2"
m2_name <- "RAV"
m2_1 <- "RAV834"
m2_2 <- "RAV833"

source("R/Fig4C_contScores.R", print.eval = TRUE)
## Warning: Removed 82 rows containing non-finite values (stat_boxplot).
## Warning: Removed 82 rows containing missing values (geom_point).

3.3 PCSS vs. RAV834/3290

RAV3290 is associated with “stage” metadata of CRC datasets.

m1_name <- "PCSS"
m1_1 <- "PCSS1"
m1_2 <- "PCSS2"
m2_name <- "RAV"
m2_1 <- "RAV834"
m2_2 <- "RAV3290"

source("R/Fig4C_contScores.R", print.eval = TRUE)
## Warning: Removed 86 rows containing non-finite values (stat_boxplot).
## Warning: Removed 86 rows containing missing values (geom_point).

3.4 PCSS vs. RAV834/596

RAV596 is associated with “grade” metadata of CRC datasets.

m1_name <- "PCSS"
m1_1 <- "PCSS1"
m1_2 <- "PCSS2"
m2_name <- "RAV"
m2_1 <- "RAV834"
m2_2 <- "RAV596"

source("R/Fig4C_contScores.R", print.eval = TRUE)
## Warning: Removed 84 rows containing non-finite values (stat_boxplot).
## Warning: Removed 84 rows containing missing values (geom_point).

4 Session Info

sessionInfo()
## R version 4.0.3 (2020-10-10)
## Platform: x86_64-pc-linux-gnu (64-bit)
## Running under: Ubuntu 18.04.5 LTS
## 
## Matrix products: default
## BLAS:   /usr/lib/x86_64-linux-gnu/openblas/libblas.so.3
## LAPACK: /usr/lib/x86_64-linux-gnu/libopenblasp-r0.2.20.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] parallel  stats     graphics  grDevices utils     datasets  methods  
## [8] base     
## 
## other attached packages:
##  [1] logistf_1.24        metafor_2.4-0       Matrix_1.3-0       
##  [4] survival_3.2-7      forcats_0.5.0       stringr_1.4.0      
##  [7] dplyr_1.0.2         purrr_0.3.4         readr_1.4.0        
## [10] tidyr_1.1.2         tibble_3.0.4        ggplot2_3.3.2      
## [13] tidyverse_1.3.0     Biobase_2.50.0      BiocGenerics_0.36.0
## [16] BiocStyle_2.18.1   
## 
## loaded via a namespace (and not attached):
##  [1] Rcpp_1.0.5           lubridate_1.7.9.2    lattice_0.20-41     
##  [4] formula.tools_1.7.1  assertthat_0.2.1     digest_0.6.27       
##  [7] R6_2.5.0             cellranger_1.1.0     backports_1.2.1     
## [10] reprex_0.3.0         evaluate_0.14        httr_1.4.2          
## [13] pillar_1.4.7         rlang_0.4.9          readxl_1.3.1        
## [16] rstudioapi_0.13      magick_2.5.2         rmarkdown_2.6       
## [19] labeling_0.4.2       splines_4.0.3        munsell_0.5.0       
## [22] broom_0.7.3          compiler_4.0.3       modelr_0.1.8        
## [25] xfun_0.19            pkgconfig_2.0.3      mgcv_1.8-33         
## [28] htmltools_0.5.0      tidyselect_1.1.0     bookdown_0.21       
## [31] fansi_0.4.1          crayon_1.3.4         dbplyr_2.0.0        
## [34] withr_2.3.0          grid_4.0.3           nlme_3.1-151        
## [37] jsonlite_1.7.2       gtable_0.3.0         lifecycle_0.2.0     
## [40] DBI_1.1.0            magrittr_2.0.1       scales_1.1.1        
## [43] cli_2.2.0            stringi_1.5.3        farver_2.0.3        
## [46] fs_1.5.0             mice_3.12.0          xml2_1.3.2          
## [49] ellipsis_0.3.1       generics_0.1.0       vctrs_0.3.6         
## [52] tools_4.0.3          glue_1.4.2           hms_0.5.3           
## [55] yaml_2.2.1           colorspace_2.0-0     BiocManager_1.30.10 
## [58] operator.tools_1.6.3 rvest_0.3.6          knitr_1.30          
## [61] haven_2.3.1

Repeat CRC paper’s Figure 4C

2021-01-06

Contents