initParallelComputing: initialise parameters for parallel computing with foreach and...

In elqumsan/RNAseqMVA: Application of Multivariate Analysis methods to RNA-seq Data

Description

by such all required paremater will be initialised for parallel computing by utilizing foreach and doParallel.

Usage

initParallelComputing(project.parameters)

Arguments

project.parameters

that are specified in yaml configuration file in oder to set all default and global variables.

Value

all project.parameters for the related recount ID.

• global: ## Global parameters are those who are the samee across all study cases

• dir: ## Directories

• main: "~/RNAseqMVA"

• workspace: "~/RNAseqMVA_workspace"

• results: "~/RNAseqMVA_workspace/results"

• TSV: "~/RNAseqMVA_workspace/data"

• memoryImages: "~/RNAseqMVA_workspace/memory_images"

• perm.prefix: "permLabels"

• jobs: "none" ## Maximal number of jobs to run in parallel. Supported: integers > 0, "auto", or "none"

• iterations: 10 ## Number of iterations for the classiifers

• reload: FALSE

• compute: TRUE ## If FALSE, do not run the heavy computations, just generate the pictures and save tables

• export.tables: TRUE ## If TRUE, save tab-delimited files with the data and results

• save.image: FALSE draw.plot : TRUE

• classifiers: "svm"

• data.types: "countsPerRun", "originalCounts", "filtered", "norm", "log2norm", "log2normPCs", "DEG", "V.importance"

• data.types.to.test: "log2normPCs"

• variables.type: "all", "top_ordered"

• deg.methods: "DESeq2", "edgeR"

• nb.variables: 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190, 200, 300, 400, 500, 1000, 2000, 5000, 10000 ## Number of variables for variable ordering test

• trainingProportion: 0.66 # ratio of spliting the data set to training and testing sets

• identicalTrainTest: TRUE

• permute: FALSE, TRUE

• verbose: TRUE

• knn:

• k: 10

• k_values: 3, 5, 7, 10, 15

• svm:

• kernel: "linear"

• kernel_values: "linear", "polynomial", "sigmoid" ## NOT WORKING YET, TO BE TESTED LATER: "radial basis"

• scale: FALSE

• type: "C-classification"

• gamma: 1

• cost: 100

• default:

• mergeRuns: TRUE ## Whether or not to merge runs per sample

• sampleIdColumn: "geo_accession"

• minSamplesPerClass: 15 ## THERE IS A BUG TO FIX: if we set this parameter to 10 (for example) the class labels must be adapted by suppressing the

• nearZeroVarFilter: TRUE

• na.rm: TRUE

Author(s)

Mustafa AbuELQumsan and Jacques van Helden.

elqumsan/RNAseqMVA documentation built on March 10, 2021, 8:10 a.m.