archR.Rmd

---
title: "archR"
author: "Irzam Sarfraz"
date: "`r Sys.Date()`"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

## Workflow

For healthy:
- Process scRNA-seq data (normalization + LSI + do clustering + find cell-types using markers + generate umap)

```{r}
library(Seurat)
library(SingleCellExperiment)
library(singleCellTK)

scRNA_healthy_1 <- readRDS("/projectnb/paxlab/isarfraz/Data/GSM4138876_scRNA_PBMC_D4T1.rds") 
seurat_healthy_rna <- CreateSeuratObject(counts = scRNA_healthy_1)
seurat_healthy_rna <- FindVariableFeatures(seurat_healthy_rna)
seurat_healthy_rna <- NormalizeData(seurat_healthy_rna)
seurat_healthy_rna <- ScaleData(seurat_healthy_rna)
# seurat_healthy_rna <- RunSLSI(seurat_healthy_rna) # this wont work because specifically designed for atac-data

#using fixed optimizeLSI code
# which input matrix (using counts)
sce <- SingleCellExperiment(list(counts=scRNA_healthy_1))
healthy_rna_lsi <- optimizeLSI(inSCE = sce, mat = scRNA_healthy_1)
embeddings <- healthy_rna_lsi$iter3$lsiObj$matSVD
healthy_rna_lsi <- healthy_rna_lsi$iter3$matNorm # sure that this mat should be used?
assay(sce, "lsi") <- healthy_rna_lsi
sce <- singleCellTK::runUMAP(inSCE = sce, useAssay = "lsi", useReducedDim = NULL, reducedDimName = "UMAP")
singleCellTK::plotUMAP(sce, reducedDimName = "UMAP")
# do clusters

# create seurat object
pca <- CreateDimReducObject(embeddings = embeddings, key = "PC_")
seurat_healthy_rna <- CreateSeuratObject(counts = assay(sce, "counts"), assay = "RNA")
seurat_healthy_rna[["pca"]] <- pca
```

- Process scATAC-seq data (gene activity scores first + LSI + do clustering + integrate with RNA for cell-types + make UMAP)
- do we need a combined UMAP for RNA and atac?



For mpal:
- do all same as above
- but create projection for healthy umap (this is not in archR - https://github.com/GreenleafLab/ArchR/discussions/1147) - can we use seurat here? yes (https://satijalab.org/seurat/articles/integration_mapping.html) (issue comes with LSI)

```{r}
scRNA_mpal_1 <- readRDS("/projectnb/paxlab/isarfraz/Data/GSM4138878_scRNA_MPAL1_T1.rds") 
sce <- SingleCellExperiment(list(counts=scRNA_mpal_1))
mpal_rna_lsi <- optimizeLSI(inSCE = sce, mat = scRNA_mpal_1)
embeddings <- mpal_rna_lsi$iter3$lsiObj$matSVD
mpal_rna_lsi <- mpal_rna_lsi$iter3$matNorm # sure that this mat should be used?
assay(sce, "lsi") <- mpal_rna_lsi
sce <- singleCellTK::runUMAP(inSCE = sce, useAssay = "lsi", useReducedDim = NULL, reducedDimName = "UMAP")
singleCellTK::plotUMAP(sce, reducedDimName = "UMAP")

pca <- CreateDimReducObject(embeddings = embeddings, key = "PC_")
seurat_mpal_rna <- CreateSeuratObject(counts = assay(sce, "counts"), assay = "RNA")
seurat_mpal_rna[["pca"]] <- pca
```

projection

```{r}
pancreas.anchors <- FindIntegrationAnchors(object.list = list(seurat_healthy_rna_1, seurat_mpal_rna), dims = 1:30)
pancreas.integrated <- IntegrateData(anchorset = pancreas.anchors, dims = 1:30)

library(ggplot2)
library(cowplot)
library(patchwork)
# switch to integrated assay. The variable features of this assay are automatically set during
# IntegrateData
DefaultAssay(pancreas.integrated) <- "integrated"
# Run the standard workflow for visualization and clustering
pancreas.integrated <- ScaleData(pancreas.integrated, verbose = FALSE)
pancreas.integrated <- RunPCA(pancreas.integrated, npcs = 30, verbose = FALSE)
pancreas.integrated <- RunUMAP(pancreas.integrated, reduction = "pca", dims = 1:30, verbose = FALSE)

tech <- c(rep("healthy", ncol(seurat_healthy_rna_1)), rep("mpal", ncol(seurat_mpal_rna)))
pancreas.integrated$tech <- tech

p1 <- DimPlot(pancreas.integrated, reduction = "umap", group.by = "tech")
p1


pancreas.query <- seurat_mpal_rna
pancreas.anchors <- FindTransferAnchors(reference = pancreas.integrated, query = pancreas.query,
    dims = 1:30, reference.reduction = "pca")
predictions <- TransferData(anchorset = pancreas.anchors, refdata = pancreas.integrated$tech,
    dims = 1:30)
pancreas.query <- AddMetaData(pancreas.query, metadata = predictions)

pancreas.integrated <- RunUMAP(pancreas.integrated, dims = 1:30, reduction = "pca", return.model = TRUE)
pancreas.query <- MapQuery(anchorset = pancreas.anchors, reference = pancreas.integrated, query = pancreas.query,
    refdata = list(tech = "tech"), reference.reduction = "pca", reduction.model = "umap")

p1 <- DimPlot(pancreas.integrated, reduction = "umap", group.by = "tech", label = TRUE, label.size = 3,
    repel = TRUE) + NoLegend() + ggtitle("Reference annotations")
p2 <- DimPlot(pancreas.query, reduction = "ref.umap", group.by = "predicted.tech", label = TRUE,
    label.size = 3, repel = TRUE) + NoLegend() + ggtitle("Query transferred labels")
p1 + p2
```


For ameya archR:
healthy
```{r}


```

mpal
```{r}
setwd("/projectnb/paxlab/isarfraz/Greenleaf_ATAC_Fragments/healthy/")
inputFiles <- list.files("/projectnb/paxlab/isarfraz/Greenleaf_ATAC_Fragments/healthy/")
ArrowFiles <- createArrowFiles(
  inputFiles = inputFiles,
  sampleNames = inputFiles,
  filterTSS = 4, #Dont set this too high because you can always increase later
  filterFrags = 1000, 
  addTileMat = TRUE,
  addGeneScoreMat = TRUE
)

doubScores <- addDoubletScores(
    input = ArrowFiles,
    k = 10, #Refers to how many cells near a "pseudo-doublet" to count.
    knnMethod = "UMAP", #Refers to the embedding to use for nearest neighbor search with doublet projection.
    LSIMethod = 1
)

projHeme1 <- ArchRProject(
  ArrowFiles = ArrowFiles, 
  outputDirectory = "HemeTutorial",
  copyArrows = TRUE #This is recommened so that if you modify the Arrow files you have an original copy for later usage.
)

projHeme1 <- addIterativeLSI(
    ArchRProj = projHeme1,
    useMatrix = "TileMatrix", 
    name = "IterativeLSI", 
    iterations = 2, 
    clusterParams = list( #See Seurat::FindClusters
        resolution = c(0.2), 
        sampleCells = 10000, 
        n.start = 10
    ), 
    varFeatures = 25000, 
    dimsToUse = 1:30
)

projHeme1 <- addClusters(
    input = projHeme1,
    reducedDims = "IterativeLSI",
    method = "Seurat",
    name = "Clusters",
    resolution = 0.8
)

projHeme1 <- addUMAP(
    ArchRProj = projHeme1, 
    reducedDims = "IterativeLSI", 
    name = "UMAP", 
    nNeighbors = 30, 
    minDist = 0.5, 
    metric = "cosine"
)

projHeme1 <- addReproduciblePeakSet(
    ArchRProj = projHeme1, 
    groupBy = "Clusters",
    peakMethod = "Tiles",
    method = "p"
)

projHeme1 <- addPeak2GeneLinks(
    ArchRProj = projHeme1,
    reducedDims = "IterativeLSI"
)
```