scFAST-seq Mutation Analysis: Data Merging, UMAP Display and Integration

Author: SeekGene

Time: 15 min

Words: 2.8k words

Updated: 2026-02-28

Reads: 0 times

scFAST-seq Notebooks Mutation Analysis

Select R environment in the top right corner

library(Seurat)
library(dplyr)
library(Matrix)
library(stringr)
library(ggplot2)
options(repr.plot.width = 12, repr.plot.height = 6)

output

The legacy packages maptools, rgdal, and rgeos, underpinning this package
will retire shortly. Please refer to R-spatial evolution reports on
https://r-spatial.org/r/2023/05/15/evolution4.html for details.
This package is now running under evolution status 0

rgeos version: 0.6-3, (SVN revision 696)
GEOS runtime version: 3.11.2-CAPI-1.17.2
Please note that rgeos will be retired during October 2023,
plan transition to sf or terra functions using GEOS at your earliest convenience.
See https://r-spatial.org/r/2023/05/15/evolution4.html for details.
GEOS using OverlayNG
Linking to sp version: 1.6-0
Polygon checking: TRUE


Attaching SeuratObject

Attaching sp


Attaching package: ‘dplyr’


The following objects are masked from ‘package:stats’:

filter, lag


The following objects are masked from ‘package:base’:

intersect, setdiff, setequal, union

Merge mutation data into single-sample rds

Input Data

1. Single-sample rds (can select rds corresponding to analysis workflow)
2. Two mutation matrices for this sample (all/alt, contact account manager to release, limited to full-length sequence data)

Matrix files can be uploaded by clicking upload in the top left corner

Content to Modify

Modify rds path below, relative directory is ../data/, absolute directory is /home/mambauser/data/

Modify mutation matrix filenames below

#load rds and matrix
obj = readRDS("data/AY1732591902625/input.rds") # AY1732591902625 is workflow ID
# AY1732591902625 is workflow ID
meta = read.table("data/AY1732591902625/meta.tsv", header = T, sep = "\t", check.names = F)
rownames(meta) = meta$barcode
obj = AddMetaData(object = obj,metadata =  meta)

snv_cover_mat = read.delim("PBMC.snp_indel.all_UMI.matrix", header = T, row.names = 1)
snv_mut_mat = read.delim("PBMC.snp_indel.alt_UMI.matrix", header = T, row.names = 1)

View Matrix

snv_cover_mat[1:3, 1:3]
snv_mut_mat[1:3, 1:3]

A data.frame: 3 × 3

	AAGTTCGTACTGGTTCT	CTGCAGGTACGGAGTAG	TAACGACCGACTGCGCA
	<int>	<int>	<int>
SDF4:chr1-1223263:T>G	0	0	0
SLC35E2B:chr1-1668373:C>T	0	0	0
CDK11A:chr1-1709071:C>T	0	0	0

A data.frame: 3 × 3

	AAGTTCGTACTGGTTCT	CTGCAGGTACGGAGTAG	TAACGACCGACTGCGCA
	<int>	<int>	<int>
SDF4:chr1-1223263:T>G	0	0	0
SLC35E2B:chr1-1668373:C>T	0	0	0
CDK11A:chr1-1709071:C>T	0	0	0

# Function：process cover/mut matrix
## object : Seurat object
## mat : matrix or dgCMatrix or data.frame
## assay_name: specified new assay name for matrix
generate_mat = function(object, mat, assay_name) {
    #make sure consistent of barcodes
    diff_barcode = setdiff(colnames(object), colnames(mat))
    if (length(diff_barcode) == ncol(object)) {
            stop("no common barcodes between matrix and seurat object")  
    }
    if (length(diff_barcode) != 0) {
        diff_matrix = matrix(0,
                         nrow = nrow(mat),
                         ncol = length(diff_barcode),
                         dimnames = list(rownames(mat), diff_barcode))
        mat = cbind(mat, diff_matrix)
    }
    mat = as(as.matrix(mat), 'dgCMatrix')
    #reorder barcode according rds
    mat = mat[, colnames(object)]
    #add new assay into rds
    object[[assay_name]] = CreateAssayObject(mat)
    return(object)
}

Run this command to add mutation matrix and coverage matrix to obj respectively

#cover matrix
obj = generate_mat(object = obj, mat = snv_cover_mat, assay_name = "SNV_all")

#mutation matrix
obj = generate_mat(object = obj, mat = snv_mut_mat, assay_name = "SNV")

output

Warning message:
“Keys should be one or more alphanumeric characters followed by an underscore, setting key from snv_all_ to snvall_”

obj has three assays, added successfully

#result
Assays(obj)

1. 'RNA'
2. 'SNV_all'
3. 'SNV'

#check value
obj@assays$SNV[1:3,1:3]

output

3 x 3 sparse Matrix of class "dgCMatrix"
AAGTTCGTACTGGTTCT CTGCAGGTACGGAGTAG CCTCTAGATGTAATTCC
SDF4:chr1-1223263:T>G . . .
SLC35E2B:chr1-1668373:C>T . . .
CDK11A:chr1-1709071:C>T . . .

Save rds

saveRDS(obj, file = "add_snv.rds")

Display Mutation Occurrence

Display Single Mutation Occurrence in Cell Population

Specify assay as SNV, i.e., specify mutation matrix, display cells with this mutation

Example mutation: SLC35E2B:chr1-1668373:C>T

DefaultAssay(obj) = "SNV"
FeaturePlot(obj, features = "SLC35E2B:chr1-1668373:C>T",ncol = 2) +
DimPlot(obj, group.by = "resolution.0.8", label = T)

Save Image

Image filename is SLC35E2B_C_to_T_mut.png, saved in current directory, can be downloaded from the left

p = FeaturePlot(obj, features = "SLC35E2B:chr1-1668373:C>T",ncol = 2) +
DimPlot(obj, group.by = "resolution.0.8", label = T)
ggsave(p, file = "SLC35E2B_C_to_T_mut.png", width = 12, height = 6)

Display Multiple Mutations Occurrence in Cell Population

Specify assay as SNV, i.e., specify mutation matrix, display cells with this mutation

Example specifying top 5 mutations

DefaultAssay(obj) = "SNV"
head5_mut = head(rownames(obj), 5)
head5_mut

1. 'SDF4:chr1-1223263:T>G'
2. 'SLC35E2B:chr1-1668373:C>T'
3. 'CDK11A:chr1-1709071:C>T'
4. 'RPL22:chr1-6197724:C>T'
5. 'DNAJC11:chr1-6667742:TTC>-'

Batch Save Images

for ( i in head5_mut) {
    p = FeaturePlot(obj, features = i, ncol = 2)
    filename = stringr::str_replace_all(i, "[:>]", "_")
    ggsave(p, file = paste0(filename, ".png"), width = 6, height = 6) # Saved in current directory
    }

Display Cell Population Covering This Site

Specify assay as SNV_all, i.e., specify coverage matrix, display cells covering the site of this mutation

DefaultAssay(obj) = "SNV_all"
FeaturePlot(obj, features = "SLC35E2B:chr1-1668373:C>T",ncol = 2) +
DimPlot(obj, group.by = "resolution.0.8", label = T)

Display Expression of Gene Containing Mutation

Specify assay as RNA, i.e., specify expression data, display expression of the gene containing this mutation

DefaultAssay(obj) = "RNA"
FeaturePlot(obj, features = "SLC35E2B",ncol = 2) +
DimPlot(obj, group.by = "resolution.0.8", label = T)

Add Mutation Status as New Label

Classify cells into two categories: Mutated/Not Mutated, and display as a new label;

Example mutation "SLC35E2B:chr1-1668373:C>T"

interest_mut = "SLC35E2B:chr1-1668373:C>T"
DefaultAssay(obj) = "SNV"
mut_label = FetchData(obj, interest_mut)
mut_label = ifelse(mut_label == 0, "negative", "positive")
table(mut_label)

output

mut_label
negative positive
1909 12

Add Label

obj = AddMetaData(obj, metadata = mut_label, col.name = "SLC35E2B_C_to_T")

Display

DefaultAssay(obj) = "RNA"
FeaturePlot(obj, features = "SLC35E2B",ncol = 2) +
DimPlot(obj, group.by = "SLC35E2B_C_to_T", cols = c("grey","red"))

Export Table

df = data.frame(barcode = colnames(obj),
                SLC35E2B_C_to_T = mut_label)
head(df)
write.csv(df, file="SLC35E2B_C_to_T_mut_meta.csv", quote=F, row.names = F)

A data.frame: 6 × 2

	barcode	SLC35E2B.chr1.1668373.C.T
	<chr>	<chr>
AAGTTCGTACTGGTTCT	AAGTTCGTACTGGTTCT	negative
CTGCAGGTACGGAGTAG	CTGCAGGTACGGAGTAG	negative
CCTCTAGATGTAATTCC	CCTCTAGATGTAATTCC	negative
TAACGACCGACTGCGCA	TAACGACCGACTGCGCA	negative
ATCAGGTGCTTTCAGAC	ATCAGGTGCTTTCAGAC	negative
GACCCTTTACGTGTTCC	GACCCTTTACGTGTTCC	negative

Upload Table to Cloud Platform

Table saved to current path, click download on the left, enter cloud platform

Cloud Platform - Select Analysis Workflow - Auxiliary Info - Label Management - Add Label - Upload Merged Meta - Click Upload

Multi-Sample Integration of Full-Length Mutation Data

Merge mutation matrices of multiple samples with integrated rds, enabling mutation correlation analysis across samples.

Input Data

1. Integrated rds. (Can select corresponding analysis workflow)
2. Mutation occurrence matrix and coverage matrix for each sample (all/alt, contact account manager to release, limited to full-length sequence data)

Example: SAMple column in integrated rds is "PBMC", mutation matrix filename is "PBMC.snp_indel.all_UMI.matrix", "PBMC.snp_indel.alt_UMI.matrix"

Matrix files can be uploaded by clicking upload in the top left corner

Content to Modify

Modify rds path below, relative directory is ../data/, absolute directory is /home/mambauser/data/

Modify mutation matrix filenames below

##read integrated rds
integrate_obj = readRDS("integrated_seurat.rds") # or integrate_obj = readRDS("data/WorkflowID/input.rds")
unique(integrate_obj$Sample)

# Matrix files for snv_cover_mat & snv_mut_mat must be in the current working directory, you can change working directory via setwd()
snv_cover_mat = c("PBMC.snp_indel.all_UMI.matrix", "cellline.snp_indel.all_UMI.matrix")
snv_mut_mat = c("PBMC.snp_indel.alt_UMI.matrix", "cellline.snp_indel.alt_UMI.matrix")

1. 'PBMC'
2. 'cellline'

# Function：process cover/mut matrix
## object : Seurat object
## mat : matrix or dgCMatrix or data.frame
## assay_name: specified new assay name for matrix
generate_mat = function(object, mat, assay_name) {
    #make sure consistent of barcodes
    diff_barcode = setdiff(colnames(object), colnames(mat))
    if (length(diff_barcode) == ncol(object)) {
            stop("no common barcodes between matrix and seurat object")  
    }
    if (length(diff_barcode) != 0) {
        diff_matrix = matrix(0,
                         nrow = nrow(mat),
                         ncol = length(diff_barcode),
                         dimnames = list(rownames(mat), diff_barcode))
        mat = cbind(mat, diff_matrix)
    }
    mat = as(as.matrix(mat), 'dgCMatrix')
    #reorder barcode according rds
    mat = mat[, colnames(object)]
    #add new assay into rds
    object[[assay_name]] = CreateAssayObject(mat)
    return(object)
}

# Function：process cover/mut matrix for integrated rds
## object: Seurat object
## snv_cover_mat: vector, cover matrix for each sample
## snv_mut_mat:vector, mut matrix for each sample
## meta_name: column name in meta.data, contains samples' name, same as marix filename's prefix
## cover_assay_name : new assay name for cover matrix
## mut_assay_name : new assay name for mut matrix
generate_mat_integrated = function(object,
                                   snv_cover_mat,
                                   snv_mut_mat,
                                   meta_name = "Sample",
                                   cover_assay_name = "SNV_all",
                                   mut_assay_name = "SNV") {

    #get barcode suffix for each sample
    suffix = as.data.frame(object@meta.data) %>%
        select(sym(meta_name)) %>%
        unique()
    suffix$barcode = rownames(suffix)
    suffix$suffix = stringr::str_replace_all(suffix$barcode,
                                             "[A-Z]{17}", "")

    #check sample name
    cover_sample = stringr::str_replace_all(snv_cover_mat,
                                            ".snp_indel.all_UMI.matrix","")
    mut_sample = stringr::str_replace_all(snv_mut_mat,
                                          ".snp_indel.alt_UMI.matrix","")
    
    if(! identical(cover_sample, mut_sample)) {
        stop("filenames are not same for snv_cover_mat and snv_mut_mat")
    }
    
    if(! all(suffix[[meta_name]] %in% cover_sample)) {
        stop("some samples do not have corresponding matrix files")
    }

    #get all matrix, add suffix
    cover_mat_list = lapply(1:nrow(suffix), function(e) {
        mat = read.delim(paste0(suffix[[meta_name]][e], ".snp_indel.all_UMI.matrix"),
                         header = T, row.names = 1)
        colnames(mat) = paste0(colnames(mat), suffix[["suffix"]][e])
        as.matrix(mat)
    })

    mut_mat_list = lapply(1:nrow(suffix), function(e) {
        mat = read.delim(paste0(suffix[[meta_name]][e], ".snp_indel.alt_UMI.matrix"),
                         header = T, row.names = 1)
        colnames(mat) = paste0(colnames(mat), suffix[["suffix"]][e])
        as.matrix(mat)
    })

    ##define two matrix
    all_muts = unique(unlist(lapply(cover_mat_list, rownames)))
    all_barcodes = unlist(lapply(cover_mat_list, colnames))
    
    all_cover_mat = Matrix::Matrix(0,
                               nrow = length(all_muts),
                               ncol = length(all_barcodes),
                               dimnames = list(all_muts, all_barcodes))
    all_mut_mat = Matrix::Matrix(0,
                               nrow = length(all_muts),
                               ncol = length(all_barcodes),
                               dimnames = list(all_muts, all_barcodes))

    #fill in value
    for (i in seq.int(cover_mat_list)) {
        mat = cover_mat_list[[i]]
        all_cover_mat[rownames(mat), colnames(mat)] = mat
        
        mat = mut_mat_list[[i]]
        all_mut_mat[rownames(mat), colnames(mat)] = mat
    }                      

    #add new assay, function is from top-line in this file
    object = generate_mat(object,
                          mat = all_cover_mat,
                          assay_name = cover_assay_name)
    
    object = generate_mat(object,
                          mat = all_mut_mat,
                          assay_name = mut_assay_name)

}

Run command to merge data

integrate_obj = generate_mat_integrated(integrate_obj,
                                        snv_cover_mat,
                                        snv_mut_mat,
                                        meta_name = "Sample")

output

Warning message:
“Keys should be one or more alphanumeric characters followed by an underscore, setting key from snv_all_ to snvall_”

New object has three assays, added successfully

Assays(integrate_obj)

1. 'RNA'
2. 'SNV_all'
3. 'SNV'

Save Data

saveRDS(integrate_obj, file = "integrate_add_snv.rds")

Can also merge data first, then integrate

#PBMC data
obj1 = readRDS("PBMC_seurat.rds")
snv_cover_mat1 = read.delim("PBMC.snp_indel.all_UMI.matrix", header = T, row.names = 1)
snv_mut_mat1 = read.delim("PBMC.snp_indel.alt_UMI.matrix", header = T, row.names = 1)
#cover matrix
obj1 = generate_mat(object = obj1, mat = snv_cover_mat1, assay_name = "SNV_all")
#mutation matrix
obj1 = generate_mat(object = obj1, mat = snv_mut_mat1, assay_name = "SNV")

output

Warning message:
“Keys should be one or more alphanumeric characters followed by an underscore, setting key from snv_all_ to snvall_”

#cellline data
obj2 = readRDS("cellline_seurat.rds")
snv_cover_mat2 = read.delim("cellline.snp_indel.all_UMI.matrix", header = T, row.names = 1)
snv_mut_mat2 = read.delim("cellline.snp_indel.alt_UMI.matrix", header = T, row.names = 1)
#cover matrix
obj2 = generate_mat(object = obj2, mat = snv_cover_mat2, assay_name = "SNV_all")
#mutation matrix
obj2 = generate_mat(object = obj2, mat = snv_mut_mat2, assay_name = "SNV")

output

Warning message:
“Keys should be one or more alphanumeric characters followed by an underscore, setting key from snv_all_ to snvall_”

integrate_obj = merge(obj1, obj2)

output

Warning message in CheckDuplicateCellNames(object.list = objects):
“Some cell names are duplicated across objects provided. Renaming to enforce unique cell names.”

New object has three assays, added successfully

Assays(integrate_obj)

1. 'RNA'
2. 'SNV_all'
3. 'SNV'

Save Data

saveRDS(integrate_obj, file = "integrate_add_snv.rds")

END