Example script demonstrating execution of the OncoSig Random Forest (OncoSig-RF) algorithm (https://pubmed.ncbi.nlm.nih.gov/32929263/), to generate an oncoprotein-specific molecular interaction map (SigMap) for the KRAS protein.
Load required libraries
if(!require("randomForest", quietly = TRUE)) {
install.packages("randomForest")
require(randomForest, quietly = TRUE)
}## randomForest 4.7-1.1## Type rfNews() to see new features/changes/bug fixes.if(!require("ROCR", quietly = TRUE)) {
install.packages("ROCR")
require(ROCR, quietly = TRUE)
}Set the working directory, where the example code and data files are located.
setwd("/usr/local/RF/")Source necessary code
source("./R/listToMatrix.R")
source("./R/OncoSigRF.R")Set the network file path. In this example script, we are using a reduced-size version of a network file generated by processing lung adenocarcinoma (luad) samples, from the Genome Cancer Atlas project (TCGA). The companion Docker container includes the full version of the network, as well as a network generated from the TCGA colon adenocarcinoma (coad) sample collection. Details about how the networks are generated can be found in the OncoSig publication cited above (PMID: 32929263).
Network_location <- "./Data_files/luad_network_sample.txt"Read in the network file and convert to matrix
Network <- read.delim(Network_location, header = F)Convert network to matrix. listToMatrix is a Josh function
Network$V1 <- as.character(Network$V1)
Network$V2 <- as.character(Network$V2)
Network$V3 <- as.numeric(Network$V3)
Network <- as.matrix(Network)
Network[, 3] <- as.numeric(Network[, 3])
Network_matrix <- listToMatrix(Network)Load the gold standard set. Here, we are using canonical KRAS signaling proteins. In additional to KRAS, gold standard sets for 9 more oncoproteins are available and can be selected by modifying the file path below, changing KRAS to the gene symbol of one of the other 9 proteins.
Gold_Standard_location <- "./Data_files/10_oncogene_pathways/KRAS/gold_standard.txt"
Gold_Standard <- read.delim(Gold_Standard_location, header = F)
Gold_Standard$V1 <- as.character(Gold_Standard$V1)The feature matrix is compiled by combining Network_matrix and Gold_Standard
Network_matrix_df <- as.data.frame(Network_matrix)
Gold_Standard_in_Network_names <- intersect(rownames(Network_matrix_df), Gold_Standard$V1)
Negative_Set_names <- setdiff(rownames(Network_matrix_df), Gold_Standard_in_Network_names)
remove(Network_matrix)Run the classifier. For this sample calculation, 10 trees and 10 iterations are sufficient.
Query_output_results <- OncoSigRF(Network_matrix_df, Gold_Standard_in_Network_names, Fraction_Gold_sample = 0.5, ntrees = 10, max_iterations = 10, balance = 1, to_save = 1)## Running OncoSig## Fraction of Gold Standard to train on for each Random Forest: 0.5## Number of Trees per Iteration: 10## Number of Iterations: 10## Balance: 1## Number of positive results to sample: 115## Number of negative results to sample: 115## Performing Random Forest## mtry equals 167## ntree OOB 1 2
## 1: 44.44% 8.82% 70.21%
## 2: 47.10% 3.17% 84.00%
## 3: 46.55% 4.76% 85.56%
## 4: 46.32% 5.38% 85.57%
## 5: 44.61% 3.92% 85.29%
## 6: 44.44% 5.77% 83.50%
## 7: 48.15% 6.54% 88.99%
## 8: 44.04% 3.70% 83.64%
## 9: 46.15% 8.26% 83.04%
## 10: 45.98% 6.31% 84.96%## Performing Random Forest
## mtry equals 167## ntree OOB 1 2
## 1: 42.70% 12.20% 68.75%
## 2: 43.07% 10.45% 74.29%
## 3: 43.45% 3.70% 80.46%
## 4: 45.60% 22.45% 69.47%
## 5: 46.41% 17.92% 75.73%
## 6: 45.87% 16.36% 75.93%
## 7: 42.99% 9.01% 77.27%
## 8: 45.54% 13.51% 76.99%
## 9: 42.41% 8.11% 76.11%
## 10: 42.67% 6.25% 78.76%## Performing Random Forest
## mtry equals 167## ntree OOB 1 2
## 1: 44.79% 8.70% 78.00%
## 2: 39.87% 2.63% 76.62%
## 3: 39.68% 3.16% 76.60%
## 4: 40.59% 1.98% 79.21%
## 5: 41.20% 1.87% 79.82%
## 6: 40.91% 1.82% 80.00%
## 7: 41.07% 1.80% 79.65%
## 8: 41.85% 0.89% 81.74%
## 9: 42.79% 2.63% 82.61%
## 10: 40.87% 1.74% 80.00%## At iteration 3## Performing Random Forest## mtry equals 167## ntree OOB 1 2
## 1: 36.26% 2.17% 71.11%
## 2: 42.75% 7.46% 76.06%
## 3: 40.46% 4.82% 73.33%
## 4: 41.18% 8.51% 74.19%
## 5: 39.90% 5.83% 73.33%
## 6: 42.79% 4.76% 79.09%
## 7: 41.26% 3.57% 79.28%
## 8: 43.75% 4.42% 83.78%
## 9: 43.11% 4.39% 82.88%
## 10: 41.15% 4.39% 78.57%## At iteration 4## Performing Random Forest## mtry equals 167## ntree OOB 1 2
## 1: 41.86% 2.38% 79.55%
## 2: 44.44% 6.06% 81.16%
## 3: 43.86% 4.65% 83.53%
## 4: 43.46% 4.21% 82.29%
## 5: 41.67% 3.85% 81.00%
## 6: 42.18% 4.72% 80.00%
## 7: 42.20% 3.64% 81.48%
## 8: 41.89% 2.65% 82.57%
## 9: 40.54% 3.54% 78.90%
## 10: 41.15% 2.63% 80.36%## At iteration 5## Performing Random Forest## mtry equals 167## ntree OOB 1 2
## 1: 50.56% 2.50% 89.80%
## 2: 42.55% 1.41% 84.29%
## 3: 46.07% 29.21% 62.92%
## 4: 46.27% 23.76% 69.00%
## 5: 44.86% 16.51% 74.29%
## 6: 41.28% 9.82% 74.53%
## 7: 39.29% 7.08% 72.07%
## 8: 40.97% 4.39% 77.88%
## 9: 41.48% 6.14% 76.52%
## 10: 40.17% 0.88% 79.13%## At iteration 6## Performing Random Forest## mtry equals 167## ntree OOB 1 2
## 1: 50.00% 17.07% 80.00%
## 2: 45.86% 9.23% 80.88%
## 3: 42.26% 6.98% 79.27%
## 4: 43.09% 6.32% 80.65%
## 5: 43.56% 5.83% 82.83%
## 6: 42.45% 4.63% 81.73%
## 7: 42.66% 3.60% 83.18%
## 8: 42.86% 4.42% 81.98%
## 9: 42.73% 3.51% 82.30%
## 10: 43.17% 4.39% 82.30%## At iteration 7## Performing Random Forest## mtry equals 167## ntree OOB 1 2
## 1: 38.67% 2.78% 71.79%
## 2: 38.97% 6.25% 68.06%
## 3: 38.73% 5.81% 71.26%
## 4: 41.18% 4.35% 76.84%
## 5: 42.79% 6.06% 78.43%
## 6: 42.06% 4.67% 79.44%
## 7: 41.01% 2.73% 80.37%
## 8: 42.60% 4.42% 81.82%
## 9: 43.17% 6.09% 81.25%
## 10: 43.61% 4.35% 83.93%## At iteration 8## Performing Random Forest## mtry equals 167## ntree OOB 1 2
## 1: 41.76% 12.20% 66.00%
## 2: 44.59% 12.68% 74.03%
## 3: 44.26% 8.99% 77.66%
## 4: 41.67% 8.82% 74.51%
## 5: 42.59% 5.45% 81.13%
## 6: 41.26% 3.54% 80.00%
## 7: 40.89% 4.39% 78.38%
## 8: 40.53% 3.48% 78.57%
## 9: 40.61% 2.61% 78.95%
## 10: 41.30% 2.61% 80.00%## At iteration 9## Performing Random Forest## mtry equals 167## ntree OOB 1 2
## 1: 37.50% 6.82% 68.18%
## 2: 40.28% 10.39% 74.63%
## 3: 44.00% 9.09% 79.31%
## 4: 43.00% 7.77% 80.41%
## 5: 42.38% 8.41% 77.67%
## 6: 42.66% 9.09% 76.85%
## 7: 41.70% 9.73% 74.55%
## 8: 42.04% 10.53% 74.11%
## 9: 44.05% 11.40% 76.99%
## 10: 42.79% 10.43% 75.44%## At iteration 10After run, save the results
result_dir <- "../result/"
if (!dir.exists(result_dir)) dir.create(result_dir, recursive = TRUE)
Query_output_results_scores <- as.data.frame(Query_output_results[[1]])Change Oncoprotein directory and output file name for other maps
write.csv(Query_output_results_scores, paste0(result_dir, "results.csv"))Create ROC curves (or save them to a PDF file, if preferred)
save_to_file = FALSE
if (save_to_file)
pdf(file = paste0(result_dir, "KRAS_ROC.pdf"))
Query_output_results_scores <- Query_output_results[[1]]
Query_output_results_scores$label <- 0
Query_output_results_scores[Gold_Standard_in_Network_names, 2] <- 1
Query_output_results_scores <- na.omit(Query_output_results_scores)
pred <- prediction(Query_output_results_scores$Score, Query_output_results_scores$label)
perf <- performance(pred, measure = "tpr", x.measure = "fpr")
plot(perf, col = "black", xlim = c(0, 1), lwd = 6)
if (save_to_file)
dev.off()Create ROC curve to FPR < 0.05
if (save_to_file)
pdf(file = paste0(result_dir, "KRAS_ROC_frp05.pdf"), height = 5, width = 5)
perf <- performance(pred, measure = "tpr", x.measure = "fpr")
plot(perf, col = "black", xlim = c(0, 0.05), lwd = 6)
if (save_to_file)
dev.off()