REWISA: REW-ISA: RNA Expression Weighted Iterative Signature...

In cst-cumt/REWISA: REW-ISA: RNA Expression Weighted Iterative Signature Algorithm

Description

REW-ISA is used to find potential local functional blocks (LFB) based on MeRIP-Seq data, where sites are hyper-methylated simultaneously across the specific conditions. REW-ISA adopts RNA expression levels of each site as weights to make sites of lower expression level less significant.

Usage

### Given the range and step size of row threshold and column threshold, optimize the selection of thresholds in the above range.
REWISA_result <- REWISA(FPKM_IP, FPKM_INPUT, optimization=TRUE, repeat_num,
                        thr_row_interval, row_step, thr_col_interval, col_step)

### Run REW-ISA under the selected optimal row and column threshold combination
REWISA_bicluster <- REWISA(FPKM_IP, FPKM_INPUT, optimization=FALSE, optimal_LFB_num, 
                           optimal_thr_row, optimal_thr_col)

Arguments

FPKM_IP	IP sample data of m6A epi-transcriptome.
FPKM_INPUT	input sample data of m6A epi-transcriptome.
MethylationLevel	Methylation level matrix.
ExpressionLevel	Expression level matrix.
optimization	Logical. If it is TRUE, start looking for the best threshold values.
optimal_thr_row	The optimal row threshold found by grid search.
optimal_thr_col	The optimal col threshold found by grid search.
optimal_LFB_num	The optimal number of LFBs found by grid search
repeat_num	The number of times to run REW-ISA repeatedly under each pair of threshold parameter settings.
thr_row_interval	Range of row threshold.
row_step	The step size of the row threshold within its range.
thr_col_interval	Range of col threshold.
col_step	The step size of the col threshold within its range.

Value

ASwC	In each repeated calculation, the Average Similarity within Clusters three-dimensional array calculated for each pair of threshold combinations.
SDwC	In each repeated calculation, the Standard Deviation within Clusters three-dimensional array calculated for each pair of threshold combinations.
LFB_num	In repeated experiments, a three-dimensional array of LFB numbers generated under each pair of threshold combinations.
ASwC_mean	The average ASwC value of each repeated calculation result in each pair of threshold combinations.
SDwC_mean	The average SDwC value of each repeated calculation result in each pair of threshold combinations.
LFB_num_mode	Under the combination of each pair of thresholds, the mode of the number of LFB is generated.
Return value	Function returns a list that stores optimized threshold combinations, the number of LFBs, or specific LFBs.

Note

Give a set of FPKM_IP and FPKM_INPUT (or MethylationLevel and ExpressionLevel) to run REW-ISA

Author(s)

Shutao Chen <[e-mail]>, Lin Zhang, Jingyi Zhu, Jia Meng and Hui Liu.

References

To use REW-ISA, please cite the following reference:

Lin Zhang, Shutao Chen, Jingyi Zhu, Jia Meng and Hui Liu. REW-ISA: unveiling local functional blocks in epi-transcriptome profiling data via an RNA expression-weighted iterative signature algorithm. BMC bioinformatics, 2020, 21(1), 1-22.

Examples

### Load data
data(Methylation_level, package = "REWISA")
data(Expression_level, package = "REWISA")

data <- as.matrix(data)
weight <- as.matrix(weight)
### Using grid search to find the optimal row threshold, column threshold and the number of LFBs
REWISA_result <- REWISA(MethylationLevel = data, ExpressionLevel = weight, 
                        optimization = TRUE, repeat_num = 40, 
                        thr_row_interval = seq(1, 3, 0.1), row_step = 0.1, 
                        thr_col_interval = seq(0.1, 1.5, 0.05), col_step = 0.05)

### The final LFBs are determined according to the optimal row threshold, column threshold and the number of LFBs.
REWISA_bicluster <- REWISA(MethylationLevel = data, ExpressionLevel = weight,
                           optimization = FALSE, optimal_LFB_num = 3, 
                           optimal_thr_row = 1.2, optimal_thr_col = 0.35)

cst-cumt/REWISA documentation built on Dec. 31, 2020, 10:10 p.m.