search

hesselberthlab / scrunchy

R toolkit for the analysis of single-cell functional heterogeneity
https://scrunchy.hesselberthlab.org/
Other
2 stars 3 forks source link

cell type classifications #47

Closed mandylr closed 5 years ago

mandylr commented 5 years ago

I used this list to make the following cell type classifications.

Markers Cell Type
IL7R CD4 T cells
CD14, LYZ CD14+ Monocytes
MS4A1 B cells
CD8A CD8 T cells
FCGR3A, MS4A7 FCGR3A+ Monocytes
GNLY, NKG7 NK cells
FCER1A, CST3 Dendritic Cells
PPBP Megakaryocytes 
library(scrunchy)
library(tidyverse)

# data from kent's newest pipeline 2019-01-03_data
dir = "/Users/mandyricher/hesselberthlab/projects/10x_haircut/20181214/data/newpipeline/"

fsce2 <- create_fsce(
        list(
                rnaseq = create_sce_rnaseq(paste(dir, "filtered_feature_bc_matrix", sep = '')),
                haircut = create_sce_haircut(paste(dir, "filtered", sep = ''), norm_method = "log_normalize")
        )
)
#> Loading sc-rnaseq matrix files: /Users/mandyricher/hesselberthlab/projects/10x_haircut/20181214/data/newpipeline/filtered_feature_bc_matrix
#> Loading haircut matrix files: /Users/mandyricher/hesselberthlab/projects/10x_haircut/20181214/data/newpipeline/filtered
# pre-calculate PCA, UMAP, t-SNE, and k-means clusters
var_genes <- calc_var_features(fsce2, "rnaseq", n = 5000)

seed = 1

fsce2 <- calc_pca(fsce2, n_pcs = 20, genes = var_genes, seed = seed)
#> scaling data
#> calculating pcs
fsce2 <- calc_umap(fsce2, n_dims = 8, seed = seed)
fsce2 <- calc_tsne(fsce2, n_dims = 8, seed = seed)
fsce2 <- calc_kmeans(fsce2, k =8, seed = seed)

fsce2_tidy <- purrr::reduce(
        list(
                tidy_dims(fsce2) %>% 
                        select(cell_id, starts_with("UMAP"), -experiment),
                tidy_coldata(fsce2),
                tidy_logcounts(fsce2[var_genes , ,"rnaseq"]) %>% 
                        select(-experiment),
                tidy_logcounts(fsce2[ , , "haircut"]) %>% 
                        select(-experiment)
        ),  
        left_join,
        by = "cell_id"
)

plot_dims(fsce2_tidy, UMAP1, UMAP2, k_cluster)


cells = c("B cells", 
          "Megakaryocytes", 
          "Monocytes", 
          "NK cells + T cells",  
          "CD4+ and CD8+ T cells", 
          "NK cells",
          "CD4+ and CD8+ T cells",
          "Dendritic cells"
)

plot_dims(fsce2_tidy, UMAP1, UMAP2, k_cluster, labels = cells)

Created on 2019-01-09 by the reprex package (v0.2.1)

jayhesselberth commented 5 years ago

These marker genes are now in pbmc_marker_genes.

library(scrunchy)
plot_dims_multi(
  fsce_tidy,
  features = pbmc_marker_genes$gene_name,
  x = UMAP1,
  y = UMAP2,
  size = 0.5
)

Created on 2019-01-15 by the reprex package (v0.2.1)

mandylr commented 5 years ago

I used this SingleR package to classify cell types and then added those cell classifications to the scrunchy objects. From my understanding SingleR will identify cell types independent of clusters (and thus takes like 2 hours to run) or it can assign cell types to clusters. It uses a reference bulk RNA seq data from isolated cell types ( It works pretty well and may be useful in determining heterogeneity within cell types. 

library(tidyverse)
#> Warning: package 'tibble' was built under R version 3.5.2
#> Warning: package 'purrr' was built under R version 3.5.2
library(scrunchy)
library(MultiAssayExperiment)
#> Loading required package: SummarizedExperiment
#> Loading required package: GenomicRanges
#> Loading required package: stats4
#> Loading required package: BiocGenerics
#> Loading required package: parallel
#> 
#> Attaching package: 'BiocGenerics'
#> The following objects are masked from 'package:parallel':
#> 
#>     clusterApply, clusterApplyLB, clusterCall, clusterEvalQ,
#>     clusterExport, clusterMap, parApply, parCapply, parLapply,
#>     parLapplyLB, parRapply, parSapply, parSapplyLB
#> The following object is masked from 'package:scrunchy':
#> 
#>     normalize
#> The following objects are masked from 'package:dplyr':
#> 
#>     combine, intersect, setdiff, union
#> The following objects are masked from 'package:stats':
#> 
#>     IQR, mad, sd, var, xtabs
#> The following objects are masked from 'package:base':
#> 
#>     anyDuplicated, append, as.data.frame, basename, cbind,
#>     colMeans, colnames, colSums, dirname, do.call, duplicated,
#>     eval, evalq, Filter, Find, get, grep, grepl, intersect,
#>     is.unsorted, lapply, lengths, Map, mapply, match, mget, order,
#>     paste, pmax, pmax.int, pmin, pmin.int, Position, rank, rbind,
#>     Reduce, rowMeans, rownames, rowSums, sapply, setdiff, sort,
#>     table, tapply, union, unique, unsplit, which, which.max,
#>     which.min
#> Loading required package: S4Vectors
#> 
#> Attaching package: 'S4Vectors'
#> The following objects are masked from 'package:dplyr':
#> 
#>     first, rename
#> The following object is masked from 'package:tidyr':
#> 
#>     expand
#> The following object is masked from 'package:base':
#> 
#>     expand.grid
#> Loading required package: IRanges
#> 
#> Attaching package: 'IRanges'
#> The following objects are masked from 'package:dplyr':
#> 
#>     collapse, desc, slice
#> The following object is masked from 'package:purrr':
#> 
#>     reduce
#> Loading required package: GenomeInfoDb
#> Loading required package: Biobase
#> Welcome to Bioconductor
#> 
#>     Vignettes contain introductory material; view with
#>     'browseVignettes()'. To cite Bioconductor, see
#>     'citation("Biobase")', and for packages 'citation("pkgname")'.
#> Loading required package: DelayedArray
#> Loading required package: matrixStats
#> 
#> Attaching package: 'matrixStats'
#> The following objects are masked from 'package:Biobase':
#> 
#>     anyMissing, rowMedians
#> The following object is masked from 'package:dplyr':
#> 
#>     count
#> Loading required package: BiocParallel
#> Warning: package 'BiocParallel' was built under R version 3.5.2
#> 
#> Attaching package: 'DelayedArray'
#> The following objects are masked from 'package:matrixStats':
#> 
#>     colMaxs, colMins, colRanges, rowMaxs, rowMins, rowRanges
#> The following object is masked from 'package:purrr':
#> 
#>     simplify
#> The following objects are masked from 'package:base':
#> 
#>     aperm, apply
library(SingleCellExperiment)
#> Warning: package 'SingleCellExperiment' was built under R version 3.5.2
source("/Users/mandyricher/src/scrunchy/R/tidiers.R")

dir = "/Users/mandyricher/hesselberthlab/projects/10x_haircut/20181214/analysis/"

#Load cell type classifications from SingleR classifications
df = read_tsv(paste0(dir, "SingleR_cell_classifications.tsv.gz"))
#> Parsed with column specification:
#> cols(
#>   cell_id = col_character(),
#>   SingleR = col_character(),
#>   HPCA1 = col_character(),
#>   HPCA2 = col_character(),
#>   Blueprint_Endcode1 = col_character(),
#>   Blueprint_Endcode2 = col_character()
#> )
df
#> # A tibble: 3,895 x 6
#>    cell_id   SingleR   HPCA1     HPCA2    Blueprint_Endco… Blueprint_Endco…
#>    <chr>     <chr>     <chr>     <chr>    <chr>            <chr>           
#>  1 AAACCTGG… CD4 T ce… T_cell:C… T_cell:… CD4+ T-cells     CD4+ T-cells    
#>  2 AAACCTGG… CD8 T ce… T_cell:C… T_cell:… CD8+ Tem         CD8+ Tem        
#>  3 AAACCTGT… CD4 T ce… T_cell:C… T_cell:… CD4+ T-cells     CD4+ T-cells    
#>  4 AAACGGGA… CD14+ Mo… Monocyte… Monocyt… Monocytes        Monocytes       
#>  5 AAACGGGA… CD4 T ce… T_cell:C… T_cell:… CD4+ T-cells     CD4+ T-cells    
#>  6 AAACGGGA… CD4 T ce… T_cell:C… T_cell:… CD4+ T-cells     CLP             
#>  7 AAACGGGC… CD4 T ce… T_cell:C… T_cell:… CD4+ T-cells     CD4+ T-cells    
#>  8 AAACGGGC… NK cells  NK_cell:… NK_cell… NK cells         NK cells        
#>  9 AAACGGGG… CD4 T ce… T_cell:C… T_cell:… CD4+ Tem         CD4+ Tem        
#> 10 AAAGATGC… NK cells  NK_cell   NK_cell… NK cells         NK cells        
#> # … with 3,885 more rows

#Load scrunchy data, cell types already added to colData using add_labels()
load(paste0(dir, "FSCE_objects_with_cell_types_from_SingleR.RData"))
colData(fsce2[['rnaseq']])
#> DataFrame with 4009 rows and 8 columns
#>                           cell_id   k_cluster         cell_type
#>                       <character> <character>       <character>
#> AAACCTGGTACCATCA AAACCTGGTACCATCA           7       CD4 T cells
#> AAACCTGGTATGAATG AAACCTGGTATGAATG           5       CD8 T cells
#> AAACCTGTCAACACGT AAACCTGTCAACACGT           7       CD4 T cells
#> AAACGGGAGAGCAATT AAACGGGAGAGCAATT           8   CD14+ Monocytes
#> AAACGGGAGTACGCGA AAACGGGAGTACGCGA           7       CD4 T cells
#> ...                           ...         ...               ...
#> TTTGTCAGTTCAGGCC TTTGTCAGTTCAGGCC           7       CD4 T cells
#> TTTGTCAGTTGTTTGG TTTGTCAGTTGTTTGG           7       CD4 T cells
#> TTTGTCAGTTTGACAC TTTGTCAGTTTGACAC           6          NK cells
#> TTTGTCATCAATACCG TTTGTCATCAATACCG           8 FCGR3A+ Monocytes
#> TTTGTCATCTGATTCT TTTGTCATCTGATTCT           4       CD4 T cells
#>                            SingleR                       HPCA1
#>                        <character>                 <character>
#> AAACCTGGTACCATCA       CD4 T cells           T_cell:CD4+_Naive
#> AAACCTGGTATGAATG       CD8 T cells T_cell:CD8+_effector_memory
#> AAACCTGTCAACACGT       CD4 T cells           T_cell:CD4+_Naive
#> AAACGGGAGAGCAATT   CD14+ Monocytes              Monocyte:CD16-
#> AAACGGGAGTACGCGA       CD4 T cells           T_cell:CD4+_Naive
#> ...                            ...                         ...
#> TTTGTCAGTTCAGGCC       CD4 T cells           T_cell:CD4+_Naive
#> TTTGTCAGTTGTTTGG       CD4 T cells           T_cell:CD4+_Naive
#> TTTGTCAGTTTGACAC          NK cells                     NK_cell
#> TTTGTCATCAATACCG FCGR3A+ Monocytes              Monocyte:CD16+
#> TTTGTCATCTGATTCT       CD4 T cells  T_cell:CD4+_central_memory
#>                                        HPCA2 Blueprint_Endcode1
#>                                  <character>        <character>
#> AAACCTGGTACCATCA           T_cell:CD4+_Naive       CD4+ T-cells
#> AAACCTGGTATGAATG T_cell:CD4+_effector_memory           CD8+ Tem
#> AAACCTGTCAACACGT           T_cell:CD4+_Naive       CD4+ T-cells
#> AAACGGGAGAGCAATT              Monocyte:CD14+          Monocytes
#> AAACGGGAGTACGCGA           T_cell:CD4+_Naive       CD4+ T-cells
#> ...                                      ...                ...
#> TTTGTCAGTTCAGGCC           T_cell:CD4+_Naive       CD4+ T-cells
#> TTTGTCAGTTGTTTGG           T_cell:CD4+_Naive       CD4+ T-cells
#> TTTGTCAGTTTGACAC                     NK_cell           NK cells
#> TTTGTCATCAATACCG              Monocyte:CD16+          Monocytes
#> TTTGTCATCTGATTCT  T_cell:CD4+_central_memory           CD4+ Tcm
#>                  Blueprint_Endcode2
#>                         <character>
#> AAACCTGGTACCATCA       CD4+ T-cells
#> AAACCTGGTATGAATG           CD8+ Tem
#> AAACCTGTCAACACGT       CD4+ T-cells
#> AAACGGGAGAGCAATT          Monocytes
#> AAACGGGAGTACGCGA       CD4+ T-cells
#> ...                             ...
#> TTTGTCAGTTCAGGCC       CD4+ T-cells
#> TTTGTCAGTTGTTTGG       CD4+ T-cells
#> TTTGTCAGTTTGACAC           NK cells
#> TTTGTCATCAATACCG          Monocytes
#> TTTGTCATCTGATTCT       CD4+ T-cells

fsce2_tidy = tidy_all(fsce2) %>%
        filter(cell_type %in% unique(df$SingleR))

plot_dims(fsce2_tidy,
          x = UMAP1, y = UMAP2,
          k_cluster)


plot_dims(fsce2_tidy,
          x = UMAP1, y = UMAP2,
          SingleR)
#> Warning: Removed 31 rows containing missing values (geom_point).


discrete_palette_default <- c(
        palette_OkabeIto_black,
        scales::brewer_pal(palette = "Paired")(12),
        scales::brewer_pal(palette = "Set1")(9),
        scales::brewer_pal(palette = "Set2")(8),
        scales::brewer_pal(palette = "Dark2")(8)
)
# Using the simple classificiations
plot_activity(fsce2_tidy, Uracil_45, SingleR) + xlim(0, 6.2) + theme(legend.position="none") 
#> Warning: Removed 31 rows containing missing values (geom_point).


# HPCA classifications
plot_activity(fsce2_tidy, Uracil_45, HPCA1) + xlim(0, 6.2) + theme(legend.position="none") + 
        scale_color_manual(values = discrete_palette_default)
#> Scale for 'colour' is already present. Adding another scale for
#> 'colour', which will replace the existing scale.


# Blueprint endcode classifications
plot_activity(fsce2_tidy, Uracil_45, Blueprint_Endcode1) + xlim(0, 6.2) + theme(legend.position="none") + 
        scale_color_manual(values = discrete_palette_default)
#> Scale for 'colour' is already present. Adding another scale for
#> 'colour', which will replace the existing scale.

Created on 2019-02-05 by the reprex package (v0.2.1)