#' @title Seurat Object.
#' @description Creating dynamic Seurat Object to add annotation information
#' and draw images.
#' @details Input takes a count, genelist and scImmuCC result, returns tsne,
#' umap, Dotplot and pheatmaps.
#' @param count  a matrix with cell unique barcodes as column names and
#' gene names as row names
#' @param genematrix Data frame with cell types as column names
#' @param ssGSEA_result Data frame, scImmuCC return result
#' @param filename Custom file name, char
#' @return 4 pictures
#' @import ggplot2
#' @import Seurat
#' @importFrom Seurat CreateSeuratObject NormalizeData FindVariableFeatures
#' ScaleData PercentageFeatureSet VariableFeatures RunPCA RunUMAP RunTSNE
#' DimPlot DotPlot DoHeatmap ProjectDim JackStraw ScoreJackStraw

seurat_Heatmap <- function(count, genematrix, ssGSEA_result, filename) {
  count <- count[, !duplicated(colnames(count))]
  celltype <- intersect(colnames(count), ssGSEA_result[, 1])
  labels <- ssGSEA_result[which(ssGSEA_result[, 1] %in% celltype), ]
  count <- count[, labels[, 1]]

  seurat.data <- CreateSeuratObject(counts = count, project = filename)
  seurat.data
  seurat.data[["percent.mt"]] <- PercentageFeatureSet(seurat.data,
    pattern = "^MT-"
  )
  HB.genes_total <- c(
    "HBA1", "HBA2", "HBB", "HBD", "HBE1", "HBG1", "HBG2",
    "HBM", "HBQ1", "HBZ"
  )
  HB_m <- match(HB.genes_total, rownames(seurat.data@assays$RNA))

  HB.genes <- rownames(seurat.data@assays$RNA)[HB_m]
  HB.genes <- HB.genes[!is.na(HB.genes)]
  seurat.data[["percent.HB"]] <- PercentageFeatureSet(seurat.data,
    features = HB.genes
  )

  seurat.data <- NormalizeData(seurat.data,
    normalization.method = "LogNormalize",
    scale.factor = 10000
  )
  seurat.data <- FindVariableFeatures(seurat.data,
    selection.method = "vst",
    nfeatures = 2000
  )
  all.genes <- rownames(seurat.data)
  seurat.data <- ScaleData(seurat.data, features = all.genes)

  # Perform linear dimensional reduction
  counts <- seurat.data[["RNA"]]$counts
  cells <- length(counts[2, ])
  if (cells > 50) {
    seurat.data <- RunPCA(seurat.data,
      features = VariableFeatures(object = seurat.data)
    )
  } else {
    seurat.data <- RunPCA(seurat.data,
      npcs = (cells - 1),
      features = VariableFeatures(object = seurat.data)
    )
  }

  head(seurat.data@reductions$pca@cell.embeddings)
  head(seurat.data@reductions$pca@feature.loadings)
  seurat.data <- ProjectDim(object = seurat.data)
  seurat.data <- JackStraw(seurat.data, num.replicate = 100)
  seurat.data <- ScoreJackStraw(seurat.data, dims = 1:20)

  seurat.data <- RunUMAP(seurat.data, dims = 1:10)
  n <- length(count[2, ])

  min_cell_count <- 100
  if (ncol(seurat.data) >= min_cell_count) {
    seurat.data <- RunTSNE(seurat.data, dims = 1:10, check_duplicates = FALSE)
  } else {
    seurat.data <- RunTSNE(seurat.data,
      dims = 1:10, perplexity = 5,
      check_duplicates = FALSE
    )
  }

  head(seurat.data@reductions$tsne@cell.embeddings)

  # Add annotation information to Seurat object
  seurat.data@meta.data$cell_type_pred <- labels[, 2]

  pdf(paste(filename, "_tSNE", ".pdf", sep = ""), width = 12, height = 10)
  p1 <- DimPlot(seurat.data,
    reduction = "tsne",
    group.by = "cell_type_pred", label = TRUE, pt.size = 1
  )
  plot(p1)
  dev.off()
  pdf(paste(filename, "_UMAP", ".pdf", sep = ""), width = 12, height = 10)
  p2 <- DimPlot(seurat.data,
    reduction = "umap",
    group.by = "cell_type_pred", label = TRUE, pt.size = 1
  )
  plot(p2)
  dev.off()

  genelist <- as.list(genematrix)
  genelist <- lapply(genelist, function(x) x[!is.na(x)])
  gene <- c()
  for (i in 1:length(genelist)) {
    gene <- c(gene, genelist[[i]])
  }

  length(gene)
  gene <- unique(gene)
  Y <- intersect(gene, rownames(count))
  length(Y)
  features <- Y
  pdf(paste(filename, "_UMAP_DotPlot", ".pdf", sep = ""),
    width = 12,
    height = 10
  )
  p3 <- DotPlot(seurat.data,
    features = features,
    group.by = "cell_type_pred"
  ) + RotatedAxis()
  plot(p3)
  dev.off()

  pdf(paste(filename, "_DoHeatmap.pdf", sep = ""), width = 12, height = 10)
  p4 <- DoHeatmap(seurat.data,
    features = Y,
    group.by = "cell_type_pred"
  ) + NoLegend()
  plot(p4)
  dev.off()
}