This question has been cross-posted on Biostars but has not received any answers yet, so I'm posting it here. I'll update here if it gets answered there.

I have downloaded a miRNA expression dataset from NCBI GEO (GSE25631) to study differential gene expression and perform other analyses. As mentioned in GEO, this profiling was performed on GPL8179 Illumina Human v2 MicroRNA expression beadchip. Accordingly, I installed the illuminaHumanv2.db package to annotate the Illumina probe IDs. However, when I tried to convert the Illumina probe IDs to their corresponding gene symbols, it returned NA for all the probes. Can anybody identify the issue with this approach? Here's the R script which I used.


Sys.setenv(VROOM_CONNECTION_SIZE = 256000)

data <- getGEO(GEO = "GSE25631", 
           destdir = "E:\\GSE25631",
           GSEMatrix = TRUE,
           AnnotGPL = FALSE,
           getGPL = FALSE,
           parseCharacteristics = TRUE)

data <- data$GSE25631_series_matrix.txt.gz

raw_intensity <- exprs(data)

samples <- as.character(pData(data)[,"title"])

raw_intensity_log <- log2(raw_intensity)
colnames(raw_intensity_log) <- c(rep("GBM", 82), rep("Normal", 5))

probe_id <- rownames(raw_intensity_log)
mapping <- data.frame(Gene = unlist(mget(x = probe_id, envir = illuminaHumanv2SYMBOL, ifnotfound = NA)))

head(mapping, 10)
ILMN_3166935   NA
ILMN_3166938   NA 
ILMN_3166940   NA
ILMN_3166941   NA 
ILMN_3166943   NA
ILMN_3166944   NA
ILMN_3166945   NA
ILMN_3166948   NA
ILMN_3166952   NA
ILMN_3166953   NA

The GSE25631_RAW.tar file in the supplementary section has all the Probe ID to Gene mappings but I'm curious to know why this approach didn't work.

  • $\begingroup$ Have you confirmed that the probe names are valid using a different tool? Could this be caused by having windows line endings in your file and those are included when querying? Or, since your E:\` path suggests you are using Windows, could it be the other way around and something is inserting a \r` before every trailing \n? $\endgroup$
    – terdon
    Apr 24, 2023 at 15:55
  • $\begingroup$ I checked the probe-to-gene mappings present in illuminav2SYMBOL. Surprisingly, none of my probe IDs are present, so it returned NAs. The list doesn't seem to contain any mappings for miRNAs. I ended up using the mapping file provided by the manufacturer. $\endgroup$
    – accibio
    Apr 25, 2023 at 6:04
  • $\begingroup$ I closed this as non-reproducible since the issue seems to be very specific to your case and your probe names. $\endgroup$
    – terdon
    Apr 25, 2023 at 11:56