rf_prediction_bracis19 Files

You can find here all script files and supplementary materials of the paper entitled "A new approach for gene prediction in prokaryotes using random forest classifier", available at BRACIS19 Files were organized into 2 folders, as follows: 
| docs: Folder which contain supplementary doc with additional information 
---| Supplementary Materials: File which contains all results by species and list of genomes used to train
| scripts: Folder which contains the scripts used to predict genes in metagenomic sequences 
---| output: Folder with all output
	---|test_rf: Folder with all test files from Random Forest
	---|prodigal: Folder with  all test files from Prodigal
	---|fraggenescan: Folder with all test files from FragGeneScan
	---|FinalModel: Generated model using Random Forest
	---|tableSequences_FinalModel.RData: Sequences used to train 
	---|tableFeatures_FinalModel.RData: Features extracted from sequences for train
---| image: Generated plots in tests 
---| fastas: Fastas files used in tests
---| input: Folder with annotated genomes used to test
---| prediction: Script to predict genes
	---|estatisticas_fraggenescan.R: Script to read FragGeneScan results
	---|estatisticas_prodigal.R: Script to read Prodigal results
	---|estatisticas_random_forest.R: Script to predict genes using Random Forest model
---| dataPreProcessing.R: Script to preprocess data 
---| extractSequences.R: Script to extract sequences 
---| functions.R: Script which contains all functions used 
---| getTableFeatures.R: Script to get a table with features from data 
---| main.R: Script to run train and test scripts 
---| modelGenerating.R: Scrit to generate models 
---| randomforestProj.Rproj: Gene prediction project 
---| friedman.R: Script to make friedman test and post-hoc