AUCell.jl

AUCell.jl is an algorithm for cell reclassification based on AUC values by feature genes in the pathway.

The algorithm is implemented in Julia. Version 1.7 or later is recommended. The simpliest way to install is using the Pkg facility in Julia.

using Pkg
Pkg.add("AUCell.jl")

Run a test job with the input files distributed with the package.

julia> using AUCell
# Use the default values for the following other parameters. If you need to modify the parameters, add them directly.
julia> result = pathway_AUC_main(use_testdata="yes")

The analysis results and a few plots will be generated and saved in the current work directory. They are also returned by the pathway_AUC_main function and can be captured by assign the returned values to a variable, e.g., result in the above example.

The first return value is a DataFrame, where rows are genes and columns are statistical values for each gene. All the genes passing the basic preprocessing step are retained.

julia> result
  1.595452 seconds (8.44 M allocations: 279.644 MiB, 4.83% gc time, 77.52% compilation time)
[ Info: INFO: The size of expression profile was (36602, 8).
  1.945127 seconds (4.95 M allocations: 260.557 MiB, 11.17% gc time, 96.92% compilation time)
[ Info: INFO: The filtered of expression profile size was (7549, 8).
  0.000401 seconds (27 allocations: 34.641 KiB)
[ Info: INFO: There are 1 pathways to be analyzed.
  0.660084 seconds (1.75 M allocations: 87.597 MiB, 3.11% gc time, 99.78% compilation time)
[ Info: INFO: According to the meta information, there are 2 groups of data and each group will be analyzed with the rest of the sample.
  2.731819 seconds (6.61 M allocations: 365.662 MiB, 3.77% gc time, 94.64% compilation time)
2×17 Matrix{Any}:
 "GeneSet"                            "group1"   "group1"   "group1"   "group1"   "group1"  …   "group2"   "group2"   "group2"   "group2"   "group2"   "group2"   "group2"   "group2"
 "HALLMARK_TNFA_SIGNALING_VIA_NFKB"  0.506962   0.500821   0.515332   0.529347   0.453294      0.506962   0.500821   0.515332   0.529347   0.453294   0.512858   0.482078   0.440029

You need to prepare two input files before the analysis: pathway features gene file and expression profile file.

1. read_gmt: Read in a GMT file (MSigDB gene set format), where fn is the file path. .gmt (See fn_feature.gmt)
2. read_gsf: Read in a general gene set file, where fn is the file path and the fields are separated by the delim character (default: white space). Each row represents a gene set and the first column is the name of the set and the rest are the genes in the set. .csv, .txt and .tsv are supported. (See .csv: fn_feature.csv or .txt: fn_feature.txt or .tsv: fn_feature.tsv)

1. read_mtx: Read in the common 10X single-cell RNA expression file in the MTX format (unzipped). (See fn: matrix.mtx, rn: features.tsv, cn: barcodes.tsv)

2. read_expr_matrix: Read in an expression matrix stored in fn where its row names are stored in rn and column names are stored in cn. (See fn: matrix.csv (.csv) or matrix.txt (.txt) or matrix.tsv (.tsv); rn: features.tsv, cn: barcodes.tsv)

read_meta: Read in a meta data file with the first row assumed to be the header and the row names assumed to be the profile names (cell barcodes). Grouping information is specified by the column with the header name of group. If group is not found, the second column will be used. It returns the grouped profile names (vector of vectors) and group names. (See fn_meta.txt)

julia> using AUCell
# Use the default values for the following other parameters. If you want to modify the parameters, add them directly.
julia> pathway_AUC_main("matrix.mtx",
                        "features.tsv",
                        "barcodes.tsv",
                     	"fn_feature.gmt",
                 		"fn_meta.txt")

Other parameters can be set by passing the value to the corresponding keyword.

pathway_AUC_main("matrix.mtx",
                 "features.tsv",
                 "barcodes.tsv",
              	 "fn_feature.gmt",
                 "fn_meta.txt";
           fn_meta_delim = '\t',
           fn_meta_group = "group",
        file_format_expr = "read_mtx",
             		   T = Int32,
             feature_col = 2,
             barcode_col = 1,
         	   rem_delim = ' ',
       feature_threshold = 30,
          cell_threshold = 200,
     file_format_feature = "read_gmt",
        fn_feature_delim = ' ',
    use_HALLMARK_pathway = "no",
                	mode = "AUCell",
           ncell_pseudo: = 0,
         auc_x_threshold = 1.0,
            remove_zeros = true,
         	use_testdata = "no",
             	work_dir = "./")

For scRNA-seq data, one can carry out a pseudobulk analysis. Rather than using the original single-cell profiles, pseudobulk profiles can be generated and used for DEG analysis. In this method, a random subset of cells from a group is aggregated into a pseudo-bulk profile.

The pseudobulk method can be turned on by setting ncell_pseudo > 0.

julia> pathway_AUC_main("matrix.mtx",
                        "features.tsv",
                        "barcodes.tsv",
                     	"fn_feature.gmt",
                 		"fn_meta.txt";
						ncell_pseudo = 10)

ncell_pseudo is the number of pseudobulk combined cells in each group. By default, profiling does not use the pseudo-bulk method (ncell_pseudo = 0). 0 indicates that the pseudo-bulk mode is not used, and other values indicate how many cells are merged into a sample.

Below lists the optional keyword parameters and their default values.

• The file content is the pathways AUC value for each group sample. Behavioral pathways, listed as samples. (See aucell_result.tsv

• AUCell_testdata.log

install.packages("JuliaCall")

library(JuliaCall)
install_julia()

library(JuliaCall)
julia <-julia_setup()

julia_install_package_if_needed("AUCell")

Run a test job with the input files distributed with the package.

julia_library("AUCell")
result <- julia_do.call("pathway_AUC_main",list(use_testdata="yes"),need_return="Julia",show_value=FALSE)

The analysis results and a few plots will be generated and saved in the current work directory. They are also returned by the pathway_AUC_main function and can be captured by assign the returned values to a variable, e.g., result in the above example.

The first return value is a DataFrame, where rows are genes and columns are statistical values for each gene. All the genes passing the basic preprocessing step are retained.

> result
  1.595452 seconds (8.44 M allocations: 279.644 MiB, 4.83% gc time, 77.52% compilation time)
[ Info: INFO: The size of expression profile was (36602, 8).
  1.945127 seconds (4.95 M allocations: 260.557 MiB, 11.17% gc time, 96.92% compilation time)
[ Info: INFO: The filtered of expression profile size was (7549, 8).
  0.000401 seconds (27 allocations: 34.641 KiB)
[ Info: INFO: There are 1 pathways to be analyzed.
  0.660084 seconds (1.75 M allocations: 87.597 MiB, 3.11% gc time, 99.78% compilation time)
[ Info: INFO: According to the meta information, there are 2 groups of data and each group will be analyzed with the rest of the sample.
  2.731819 seconds (6.61 M allocations: 365.662 MiB, 3.77% gc time, 94.64% compilation time)
2×17 Matrix{Any}:
 "GeneSet"                            "group1"   "group1"   "group1"   "group1"   "group1"  …   "group2"   "group2"   "group2"   "group2"   "group2"   "group2"   "group2"   "group2"
 "HALLMARK_TNFA_SIGNALING_VIA_NFKB"  0.506962   0.500821   0.515332   0.529347   0.453294      0.506962   0.500821   0.515332   0.529347   0.453294   0.512858   0.482078   0.440029

You need to prepare four input files before the analysis: metadata file and expression matrix. You need to prepare two input files

1. read_gmt: Read in a GMT file (MSigDB gene set format), where fnis the file path..gmt` (See fn_feature.gmt)
2. read_gsf: Read in a general gene set file, where fn is the file path and the fields are separated by the delim character (default: white space). Each row represents a gene set and the first column is the name of the set and the rest are the genes in the set. .csv, .txt and .tsv are supported. (See .csv: fn_feature.csv or .txt: fn_feature.txt or .tsv: fn_feature.tsv)

1. read_mtx: Read in the common 10X single-cell RNA expression file in the MTX format (unzipped). (See fn: matrix.mtx, rn: features.tsv, cn: barcodes.tsv)

2. read_expr_matrix: Read in an expression matrix stored in fn where its row names are stored in rn and column names are stored in cn. (See fn: matrix.csv (.csv) or matrix.txt (.txt) or matrix.tsv (.tsv); rn: features.tsv, cn: barcodes.tsv)

read_meta: Read in a meta data file with the first row assumed to be the header and the row names assumed to be the profile names (cell barcodes). Grouping information is specified by the column with the header name of group. If group is not found, the second column will be used. It returns the grouped profile names (vector of vectors) and group names. (See fn_meta.txt)

Once the files are ready, you can carry out the AUCell analysis with the default settings as follows.

julia_library("AUCell")
# Use the default values for the following other parameters. If you want to modify the parameters, add them directly.
julia_do.call("reoa",list("matrix.mtx",
                         "features.tsv",
                         "barcodes.tsv",
                         "fn_feature.gmt",
                         "fn_meta.txt"),need_return="Julia",show_value=FALSE)

Other parameters can be set by passing the value to the corresponding keyword.

julia_do.call("reoa",list("matrix.mtx",
                         "features.tsv",
                         "barcodes.tsv",
                         "fn_feature.gmt",
                         "fn_meta.txt";
                   fn_meta_delim = '\t',
                   fn_meta_group = "group",
                file_format_expr = "read_mtx",
                               T = Int32,
                     feature_col = 2,
                     barcode_col = 1,
         			   rem_delim = ' ',
               feature_threshold = 30,
                  cell_threshold = 200,
             file_format_feature = "read_gmt",
                fn_feature_delim = ' ',
            use_HALLMARK_pathway = "no",
                            mode = "AUCell",
                   ncell_pseudo: = 0,
                 auc_x_threshold = 1.0,
                    remove_zeros = true,
                    use_testdata = "no",
                        work_dir = "./"),need_return="Julia",show_value=FALSE)

For scRNA-seq data, one can carry out a pseudobulk analysis. Rather than using the original single-cell profiles, pseudobulk profiles can be generated and used for DEG analysis. In this method, a random subset of cells from a group is aggregated into a pseudo-bulk profile.

The pseudobulk method can be turned on by setting ncell_pseudo > 0.

julia_do.call("reoa",list("matrix.mtx",
                        "features.tsv",
                        "barcodes.tsv",
                     	"fn_feature.gmt",
                 		"fn_meta.txt";
						ncell_pseudo = 10),need_return="Julia",show_value=FALSE)

ncell_pseudo is the number of pseudobulk combined cells in each group. By default, profiling does not use the pseudo-bulk method (ncell_pseudo = 0). 0 indicates that the pseudo-bulk mode is not used, and other values indicate how many cells are merged into a sample.

See 1.3 Optional Parameters.

See 1.4 Example output file.