assign.preprocess: Input data preprocessing
In ASSIGN: Adaptive Signature Selection and InteGratioN (ASSIGN)

Description Usage Arguments Details Value Author(s) Examples

The assign.preprocess function is used to perform quality control on the user-provided input data and generate starting values and/or prior values for the model parameters. The assign.preprocess function is optional. For users who already have the correct format for the input of the assign function, they can skip this step and go directly to the assign.mcmc function.

assign.preprocess(
  trainingData = NULL,
  testData,
  anchorGenes = NULL,
  excludeGenes = NULL,
  trainingLabel,
  geneList = NULL,
  n_sigGene = NA,
  theta0 = 0.05,
  theta1 = 0.9,
  pctUp = 0.5,
  geneselect_iter = 500,
  geneselect_burn_in = 100,
  progress_bar = TRUE
)

`trainingData`	The genomic measure matrix of training samples (i.g., gene expression matrix). The dimension of this matrix is probe number x sample number. The default is NULL.
`testData`	The genomic measure matrix of test samples (i.g., gene expression matrix). The dimension of this matrix is probe number x sample number.
`anchorGenes`	A list of genes that will be included in the signature even if they are not chosen during gene selection.
`excludeGenes`	A list of genes that will be excluded from the signature even if they are chosen during gene selection.
`trainingLabel`	The list linking the index of each training sample to a specific group it belongs to. See details and examples for more information.
`geneList`	The list that collects the signature genes of one/multiple pathways. Every component of this list contains the signature genes associated with one pathway. The default is NULL.
`n_sigGene`	The vector of the signature genes to be identified for one pathway. n_sigGene needs to be specified when geneList is set NULL. The default is NA. See examples for more information.
`theta0`	The prior probability for a gene to be significant, given that the gene is NOT defined as "significant" in the signature gene lists provided by the user. The default is 0.05.
`theta1`	The prior probability for a gene to be significant, given that the gene is defined as "significant" in the signature gene lists provided by the user. The default is 0.9.
`pctUp`	By default, ASSIGN bayesian gene selection chooses the signature genes with an equal fraction of genes that increase with pathway activity and genes that decrease with pathway activity. Use the pctUp parameter to modify this fraction. Set pctUP to NULL to select the most significant genes, regardless of direction. The default is 0.5
`geneselect_iter`	The number of iterations for bayesian gene selection. The default is 500.
`geneselect_burn_in`	The number of burn-in iterations for bayesian gene selection. The default is 100
`progress_bar`	Display a progress bar for gene selection. Default is TRUE.

The assign.preprocess is applied to perform quality control on the user-provided genomic data and meta data, re-format the data in a way that can be used in the following analysis, and generate starting/prior values for the pathway signature matrix. The output values of the assign.preprocess function will be used as input values for the assign.mcmc function.

For training data with 1 control group and 3 experimental groups (10 samples/group; all 3 experimental groups share 1 control group), the trainingLabel can be specified as: trainingLabel <- list(control = list(expr1=1:10, expr2=1:10, expr3=1:10), expr1 = 11:20, expr2 = 21:30, expr3 = 31:40)

For training data with 3 control groups and 3 experimental groups (10 samples/group; Each experimental group has its corresponding control group), the trainingLabel can be specified as: trainingLabel <- list(control = list(expr1=1:10, expr2=21:30, expr3=41:50), expr1 = 11:20, expr2 = 31:40, expr3 = 51:60)

It is highly recommended that the user use the same experiment name when specifying control indices and experimental indices.

`trainingData_sub`	The G x N matrix of G genomic measures (i.g., gene expression) of N training samples. Genes/probes present in at least one pathway signature are retained. Only returned when the training dataset is available.
`testData_sub`	The G x N matrix of G genomic measures (i.g., gene expression) of N test samples. Genes/probes present in at least one pathway signature are retained.
`B_vector`	The G x 1 vector of genomic measures of the baseline/background. Each element of the B_vector is calculated as the mean of the genomic measures of the control samples in training data.
`S_matrix`	The G x K matrix of genomic measures of the signature. Each column of the S_matrix represents a pathway. Each element of the S_matrix is calculated as the mean of genomic measures of the experimental samples minus the mean of the control samples in the training data.
`Delta_matrix`	The G x K matrix of binary indicators. Each column of the Delta_matrix represents a pathway. The elements in Delta_matrix are binary (0, insignificant gene; 1, significant gene).
`Pi_matrix`	The G x K matrix of probability p of a Bernoulli distribution. Each column of the Pi_matrix represents a pathway. Each element in the Pi_matrix is the probability of a gene to be significant in its associated pathway.
`diffGeneList`	The list that collects the signature genes of one/multiple pathways generated from the training samples or from the user provided gene list. Every component of this list contains the signature genes associated with one pathway.

Ying Shen

processed.data <- assign.preprocess(trainingData=trainingData1,
                                    testData=testData1,
                                    trainingLabel=trainingLabel1,
                                    geneList=geneList1)