Wrapping up a new list of projection matrices with the list computed from the function of sjdWrap
Usage
sjdWrapProjection(
data.list.template,
data.list.projection,
species.template,
species.vec.projection,
geneType.template,
geneType.vec.projection
)Arguments
- data.list.template
input list of expression matrices from the output of the sjdWrap function
- data.list.projection
input list of expression matrices needed to be matched with data.list.template
- species.template
character of species type from the sjdWrap function
- species.vec.projection
list of species of each matrix to be projected
- geneType.template
character of gene/rowname type from the sjdWrap function
- geneType.vec.projection
list of output gene/rowname type of each matrix to be projected
Examples
## Load NeuroGenesis4 data into R
data(NeuroGenesis4)
## sjdWrap of the training data sets
SJDdataIN = sjdWrap(
data.list = NeuroGenesis4,
species.vector=c("human","human","human","mouse"),
geneType.vector=c("symbol","ensembl","symbol","symbol"),
geneType.out="symbol",
species.out="human")
#> Using biomaRt to connect gene IDs across 4 datasets:
#> Getting biomaRt IDs for dataset 1
#> You have input 40 genes
#> We found 40 matches
#> 1 of those are duplicates and only keeping the 1st of each
#> Getting biomaRt IDs for dataset 2
#> You have input 40 genes
#> We found 37 matches
#> 1 of those are duplicates and only keeping the 1st of each
#> Getting biomaRt IDs for dataset 3
#> You have input 40 genes
#> We found 39 matches
#> 0 of those are duplicates and only keeping the 1st of each
#> Getting biomaRt IDs for dataset 4
#> You have input 40 genes
#> We found 31 matches
#> 0 of those are duplicates and only keeping the 1st of each
#> constructed 4 tables of cross-species matching genes
#> we found 24 shared genes in 4 datasets
#> new data list of 4 datasets constructed
## Sample from data, serving as the projection expression matrices
NeuroGenesis4.sample = NeuroGenesis4
NeuroGenesis4.sample[[1]] = NeuroGenesis4.sample[[1]][-5,]
rownames(NeuroGenesis4.sample[[1]])[5] = paste0(rownames(NeuroGenesis4.sample[[1]])[5], ".test")
NeuroGenesis4.sample[[2]] = NeuroGenesis4.sample[[2]][-10,]
rownames(NeuroGenesis4.sample[[2]])[10] = paste0(rownames(NeuroGenesis4.sample[[2]])[10], ".test")
NeuroGenesis4.sample[[3]] = NeuroGenesis4.sample[[3]][-15,]
rownames(NeuroGenesis4.sample[[3]])[15] = paste0(rownames(NeuroGenesis4.sample[[3]])[15], ".test")
NeuroGenesis4.sample[[4]] = NeuroGenesis4.sample[[4]][-20,]
rownames(NeuroGenesis4.sample[[4]])[20] = paste0(rownames(NeuroGenesis4.sample[[4]])[20], ".test")
SJDdataProjection = sjdWrapProjection(
SJDdataIN, NeuroGenesis4.sample, "human", c("human","human","human","mouse"),
"symbol", c("symbol","ensembl","symbol","symbol"))
#> Using biomaRt to connect gene IDs across 4 datasets:
#> Getting biomaRt IDs for dataset 1
#> You have input 39 genes
#> We found 38 matches
#> 1 of those are duplicates and only keeping the 1st of each
#> Getting biomaRt IDs for dataset 2
#> You have input 39 genes
#> We found 35 matches
#> 1 of those are duplicates and only keeping the 1st of each
#> Getting biomaRt IDs for dataset 3
#> You have input 39 genes
#> We found 37 matches
#> 0 of those are duplicates and only keeping the 1st of each
#> Getting biomaRt IDs for dataset 4
#> You have input 39 genes
#> We found 29 matches
#> 0 of those are duplicates and only keeping the 1st of each
#> constructed 4 tables of cross-species matching genes