Errors and Warnings with ImmuCellAI_new and ImmuCellAI functions when requesting ICB response predictions

[s-bell]

Hello,

Thank you for putting together this package. Unfortunately I've encountered a few issues when attempting to use it - particularly when invoking the option to calculate ICB response predictions.

Issue description:
A series of errors and warnings while using both ImmuCellAI_new and ImmuCellAI functions. The issues range from missing objects to type coercion warnings and deprecated array recycling. This happens when the input is a filtered matrix.

Reproducible example:

# Read the data into a data frame
data_df <- read.table("http://bioinfo.life.hust.edu.cn/static/ImmuCellAI/data/AML_sample.txt", header = TRUE, sep = "\t", stringsAsFactors = FALSE)

# Convert the data frame to a matrix
data_matrix <- as.matrix(data_df)

# Set row and column names
rownames(data_matrix) <- data_matrix[,1]
colnames(data_matrix) <- names(data_df)

# Remove the first column as it's now the row names
data_matrix <- data_matrix[,-1]

# Identify columns that start with a specific letter, e.g., 'X'
cols_to_remove <- grepl("^X", colnames(data_matrix))

# Remove those columns
data_matrix <- data_matrix[, !cols_to_remove]

# Filter out genes with constant expression across all samples
non_constant_genes <- which(apply(data_matrix, 1, function(x) length(unique(x)) > 1))
filtered_expression_matrix <- data_matrix[non_constant_genes, ]

# Try analysis
test <- ImmuCellAI_new(filtered_expression_matrix, "rnaseq", 0, 1)

# Setting parallel calculations through a MulticoreParam back-end
# with workers=20 and tasks=100.
# Estimating ssGSEA scores for 10 gene sets.
# [1] "Calculating ranks..."
# [1] "Calculating absolute values from ranks..."
# |=========================================================================================| 100%
# 
# [1] "Normalizing..."
# Setting parallel calculations through a MulticoreParam back-end
# with workers=20 and tasks=100.
# Estimating ssGSEA scores for 14 gene sets.
# [1] "Calculating ranks..."
# [1] "Calculating absolute values from ranks..."
# |=========================================================================================| 100%
# 
# [1] "Normalizing..."
# Error in result_layer(t(round(all_norm, 3)), group_tag, response_tag) : 
#   object 'train_data' not found
# In addition: Warning message:
#   In .filterFeatures(expr, method) :
#   1 genes with constant expression values throuhgout the samples.

# Turn off ICM

test2 <- ImmuCellAI_new(filtered_expression_matrix, "rnaseq",0,0)

# Setting parallel calculations through a MulticoreParam back-end
# with workers=20 and tasks=100.
# Estimating ssGSEA scores for 10 gene sets.
# [1] "Calculating ranks..."
# [1] "Calculating absolute values from ranks..."
# |=========================================================================================| 100%
# 
# [1] "Normalizing..."
# Setting parallel calculations through a MulticoreParam back-end
# with workers=20 and tasks=100.
# Estimating ssGSEA scores for 14 gene sets.
# [1] "Calculating ranks..."
# [1] "Calculating absolute values from ranks..."
# |=========================================================================================| 100%
# 
# [1] "Normalizing..."
# Warning message:
#   In .filterFeatures(expr, method) :
#   1 genes with constant expression values throuhgout the samples.

# Find genes with constant expression across all samples
constant_gene <- which(apply(filtered_expression_matrix, 1, function(x) length(unique(x)) == 1))

# Get the names of these genes
constant_gene_name <- rownames(filtered_expression_matrix)[constant_gene]

# Report the genes
print(constant_gene_name)
# character(0)

# Training data not loaded with ImmuCellAI_new function (https://github.com/lydiaMyr/ImmuCellAI/blob/main/R/ImmuCellAI.R - beginning line 317)
data("train_data")
data("train_tag")

test3 <- ImmuCellAI_new(filtered_expression_matrix, "rnaseq", 0, 1)

# Setting parallel calculations through a MulticoreParam back-end
# with workers=20 and tasks=100.
# Estimating ssGSEA scores for 10 gene sets.
# [1] "Calculating ranks..."
# [1] "Calculating absolute values from ranks..."
# |=========================================================================================| 100%
# 
# [1] "Normalizing..."
# Setting parallel calculations through a MulticoreParam back-end
# with workers=20 and tasks=100.
# Estimating ssGSEA scores for 14 gene sets.
# [1] "Calculating ranks..."
# [1] "Calculating absolute values from ranks..."
# |=========================================================================================| 100%
# 
# [1] "Normalizing..."
# Error in if (any(as.integer(y) != y)) stop("dependent variable has to be of factor or integer type for classification mode.") : 
#   missing value where TRUE/FALSE needed
# In addition: Warning messages:
#   1: In .filterFeatures(expr, method) :
#   1 genes with constant expression values throuhgout the samples.
# 2: In cret$cresults * scale.factor :
#   Recycling array of length 1 in vector-array arithmetic is deprecated.
# Use c() or as.vector() instead.
# 
# 3: In svm.default(x, y, scale = scale, ..., na.action = na.action) :
#   NAs introduced by coercion

# Use original function
test4 <- ImmuCellAI(filtered_expression_matrix, "rnaseq", 0, 1, 0)

# Setting parallel calculations through a MulticoreParam back-end
# with workers=20 and tasks=100.
# Estimating ssGSEA scores for 24 gene sets.
# [1] "Calculating ranks..."
# [1] "Calculating absolute values from ranks..."
# |=========================================================================================| 100%
# 
# [1] "Normalizing..."
# Error in if (any(as.integer(y) != y)) stop("dependent variable has to be of factor or integer type for classification mode.") : 
#   missing value where TRUE/FALSE needed
# In addition: Warning messages:
#   1: In .filterFeatures(expr, method) :
#   1 genes with constant expression values throuhgout the samples.
# 2: In cret$cresults * scale.factor :
#   Recycling array of length 1 in vector-array arithmetic is deprecated.
# Use c() or as.vector() instead.
# 
# 3: In svm.default(x, y, scale = scale, ..., na.action = na.action) :
#   NAs introduced by coercion

# Convert train tag to a dataframe and set as factor
train_tag <- as.data.frame(train_tag, stringsAsFactors = TRUE)

# Use new function
test5 <- ImmuCellAI_new(filtered_expression_matrix, "rnaseq", 0, 1)

# Setting parallel calculations through a MulticoreParam back-end
# with workers=20 and tasks=100.
# Estimating ssGSEA scores for 10 gene sets.
# [1] "Calculating ranks..."
# [1] "Calculating absolute values from ranks..."
# |=========================================================================================| 100%
# 
# [1] "Normalizing..."
# Setting parallel calculations through a MulticoreParam back-end
# with workers=20 and tasks=100.
# Estimating ssGSEA scores for 14 gene sets.
# [1] "Calculating ranks..."
# [1] "Calculating absolute values from ranks..."
# |=========================================================================================| 100%
# 
# [1] "Normalizing..."
# Warning messages:
#   1: In .filterFeatures(expr, method) :
#   1 genes with constant expression values throuhgout the samples.
# 2: In cret$cresults * scale.factor :
#   Recycling array of length 1 in vector-array arithmetic is deprecated.
# Use c() or as.vector() instead.

# Sanity check for original function
test6 <- ImmuCellAI(filtered_expression_matrix, "rnaseq", 0, 1, 0)

# Setting parallel calculations through a MulticoreParam back-end
# with workers=20 and tasks=100.
# Estimating ssGSEA scores for 24 gene sets.
# [1] "Calculating ranks..."
# [1] "Calculating absolute values from ranks..."
# |=========================================================================================| 100%
# 
# [1] "Normalizing..."
# Error in if (any(as.integer(y) != y)) stop("dependent variable has to be of factor or integer type for classification mode.") : 
#   missing value where TRUE/FALSE needed
# In addition: Warning messages:
#   1: In .filterFeatures(expr, method) :
#   1 genes with constant expression values throuhgout the samples.
# 2: In cret$cresults * scale.factor :
#   Recycling array of length 1 in vector-array arithmetic is deprecated.
# Use c() or as.vector() instead.
# 
# 3: In svm.default(x, y, scale = scale, ..., na.action = na.action) :
#   NAs introduced by coercion

# Reads train_tag as a series of 'R' and 'NR' characters

Expected behaviour:
The function should have been able to handle the filtered matrix without errors or warnings.

Actual behaviour:

1. Error in result_layer: object 'train_data' not found.
2. Warning: 1 genes with constant expression values throughout the samples.
3. Error: missing value where TRUE/FALSE needed.
4. Warning: Recycling array of length 1 in vector-array arithmetic is deprecated.

Environment:

> sessionInfo()
R version 4.3.1 (2023-06-16)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 22.04.3 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3 
LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so;  LAPACK version 3.10.0

locale:
 [1] LC_CTYPE=en_GB.UTF-8       LC_NUMERIC=C               LC_TIME=en_GB.UTF-8       
 [4] LC_COLLATE=en_GB.UTF-8     LC_MONETARY=en_GB.UTF-8    LC_MESSAGES=en_GB.UTF-8   
 [7] LC_PAPER=en_GB.UTF-8       LC_NAME=C                  LC_ADDRESS=C              
[10] LC_TELEPHONE=C             LC_MEASUREMENT=en_GB.UTF-8 LC_IDENTIFICATION=C       

time zone: Europe/London
tzcode source: system (glibc)

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] ImmuCellAI_0.1.0 quadprog_1.5-8   pracma_2.4.2     e1071_1.7-13     gridExtra_2.3   
[6] GSVA_1.48.3      devtools_2.4.5   usethis_2.2.2   

Thank you for your assistance with this issue.

All the best,

Steven.