Goolam

Description

Goolam dataset in list format, include scRNA-seq data and cell type information.

Usage

Goolam

Format

An object of class list of length 2.

Goolam_result

Description

Result of processing Goolam dataset using 'scDHA' function.

Usage

Goolam_result

Format

An object of class list of length 4.

scDHA

Description

The main function to perform dimension deduction and clustering.

Usage

scDHA(
  data = data,
  k = NULL,
  method = "scDHA",
  sparse = FALSE,
  n = 5000,
  ncores = 10L,
  gen_fil = TRUE,
  do.clus = TRUE,
  sample.prob = NULL,
  seed = NULL
)

Arguments

Value

List with the following keys:

• cluster - A numeric vector containing cluster assignment for each sample. If do.clus = False, this values is always NULL.

• latent - A matrix representing compressed data from the input data, with rows represent samples and columns represent latent variables.

Examples

library(scDHA)
#Load example data (Goolam dataset)
data('Goolam'); data <- t(Goolam$data); label <- as.character(Goolam$label)
#Log transform the data 
data <- log2(data + 1)
if(torch::torch_is_installed()) #scDHA need libtorch installed
{
  #Generate clustering result, the input matrix has rows as samples and columns as genes
  result <- scDHA(data, ncores = 2, seed = 1)
  #The clustering result can be found here 
  cluster <- result$cluster
}

scDHA classification

Description

Perform classification of new data based on available data.

Usage

scDHA.class(
  train = train,
  train.label = train.label,
  test = test,
  ncores = 10L,
  seed = NULL
)

Arguments

Value

A vector contain classified labels for new data.

Examples

library(scDHA)
#Load example data (Goolam dataset)
data('Goolam'); data <- t(Goolam$data); label <- as.character(Goolam$label)
#Log transform the data 
data <- log2(data + 1)
#Split data into training and testing sets
set.seed(1)
idx <- sample.int(nrow(data), size = round(nrow(data)*0.75))
train.x <- data[idx, ]; train.y <- label[idx]
test.x <- data[-idx, ]; test.y <- label[-idx]
if(torch::torch_is_installed()) #scDHA need libtorch installed
{
  #Predict the labels of cells in testing set
  prediction <- scDHA.class(train = train.x, train.label = train.y, test = test.x, 
                            ncores = 2, seed = 1)
  #Calculate accuracy of the predictions
  acc <- round(sum(test.y == prediction)/length(test.y), 2)
  print(paste0("Accuracy = ", acc))
}

scDHA pseudo time inference

Description

Inferring pseudo-time data.

Usage

scDHA.pt(sc = sc, start.point = 1, ncores = 10L, seed = NULL)

Arguments

Value

List with the following keys:

• pt - Pseudo-time values for each sample.

Examples

library(scDHA)
#Load example data (Goolam dataset)
data('Goolam'); data <- t(Goolam$data); label <- as.character(Goolam$label)
#Log transform the data 
data <- log2(data + 1)
if(torch::torch_is_installed()) #scDHA need libtorch installed
{
  #Generate clustering result, the input matrix has rows as samples and columns as genes
  result <- scDHA(data, ncores = 2, seed = 1)
  #Cell stage order in Goolam dataset
  cell.stages <- c("2cell", "4cell", "8cell", "16cell", "blast")
  #Generate pseudo-time for each cell, the input is the output from scDHA function
  result <- scDHA.pt(result, start.point = 1, ncores = 2, seed = 1)
  #Calculate R-squared value 
  r2 <- round(cor(result$pt, as.numeric(factor(label, levels = cell.stages)))^2, digits = 2)
}

scDHA visulization

Description

Generating 2D embeded data for visulation.

Usage

scDHA.vis(sc = sc, method = "UMAP", ncores = 10L, seed = NULL)

Arguments

Value

a list with the following keys:

• pred - A matrix representing the 2D projection of single-cell data, where rows represent samples and columns represent latent components.

Examples

library(scDHA)
#Load example data (Goolam dataset)
data('Goolam'); data <- t(Goolam$data); label <- as.character(Goolam$label)
#Log transform the data 
data <- log2(data + 1)
if(torch::torch_is_installed()) #scDHA need libtorch installed
{
  #Generate clustering result, the input matrix has rows as samples and columns as genes
  result <- scDHA(data, ncores = 2, seed = 1)
  #Generate 2D representation, the input is the output from scDHA function
  result <- scDHA.vis(result, ncores = 2, seed = 1)
  #Plot the representation of the dataset, different colors represent different cell types
  plot(result$pred, col=factor(label), xlab = "scDHA1", ylab = "scDHA2")
}