Cannot read a specific FCS 3.1 file for unknown reason

[CroixJeremy2]

Hello,
I am quite new in reading fcs files using R, therefore I am sorry in advance if the issue have already been raised in the past (I couldn't find similar issues on RGLab/flowCore/issues though).

Here are three FCS 3.1 files that have been generated on the same machine (https://research.pasteur.fr/en/equipment/big-foot/) during the same day for a sorting experiment: https://dl.pasteur.fr/fop/P67e8o6r/Test_folder.zip

However, I cannot read Sample_A, while Sample_B and Sample_C can be perfectly read using read.FSC(). Interestingly, all three files are correctly detected as FCS files using isFCSfile(). The problem happens in MacOS and also on Windows with fresh installation of R and flowCore. Here are the script and outputs from R:

library(flowCore)

# Windows
# setwd("C:/Users/jchantre/Desktop/Test_folder")

# MacOS
# setwd("/Users/jchantre/Desktop/Test_folder")

isFCSfile("Sample_A.fcs")
isFCSfile("Sample_B.fcs")
isFCSfile("Sample_C.fcs")

a = read.FCS("Sample_A.fcs")
b = read.FCS("Sample_B.fcs")
c = read.FCS("Sample_C.fcs")

head(a)
head(b)
head(c)

On MacOS:

On Windows:

Is there a way to solve this error message?

> a = read.FCS("Sample_A.fcs")
Error in rawToChar(txt) : 
  embedded nul in string: '*$BEGINANALYSIS*000000000000*$BEGINDATA*000000006769*$BEGINSTEXT*000000000000*$BYTEORD*1,2,3,4*$DATATYPE*F*$ENDANALYSIS*000000000000*$ENDDATA*005622116501*$ENDSTEXT*000000000000*$MODE*L*$NEXTDATA*000000000000*$PAR*29*$P1B*32*$P1E*0,0*$P1N*Time*$P1R*2147483647*$P1L*0*$P1O*0*$P1S*Time*$P1V*0*$P2B*32*$P2E*0,0*$P2N*FSC04-A*$P2R*100000*$P2F*488_FSC*$P2L*488*$P2O*125*$P2S*488 FSC-A*$P2V*306*$P3B*32*$P3E*0,0*$P3N*SSC56-A*$P3R*100000*$P3F*488_SSC*$P3L*488*$P3O*125*$P3S*488 SSC-A*$P3V*362*$P4B*32*$P4E*0,0*$P4N*FSC04-H*$P4R*100000*$P4F*488_FSC*$P4L*488*$P4O*125*$P4S*488 FSC-H*$P4V*306*$P5B*32*$P5E*0,0*$P5N*FSC04-W*$P5R*100000*$P5F*488_FSC*$P5L*488*$P5O*125*$P5S*488 FSC-W*$P5V*306*$P6B*32*$P6E*0,0*$P6N*SSC56-H*$P6R*100000*$P6F*488_SSC*$P6L*488*$P6O*125*$P6S*488 SSC-H*$P6V*362*$P7B*32*$P7E*0,0*$P7N*SSC56-W*$P7R*100000*$P7F*488_SSC*$P7L*488*$P7O*125*$P7S*488 SSC-W*$P7V*362*$P8B*32*$P8E*0,0*$P8N*FSC58-H*$P8R*100000*$P8F*488_FSC_Polar*$P8L*488*$P8O*125*$P8S*488 FSC P

Thanks in advance for your response,
Best regards,

sessionInfo() in MacOS:

R version 4.3.3 (2024-02-29) Platform: x86_64-apple-darwin20 (64-bit) Running under: macOS Monterey 12.7.5

Matrix products: default
BLAS: /Library/Frameworks/R.framework/Versions/4.3-x86_64/Resources/lib/libRblas.0.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/4.3-x86_64/Resources/lib/libRlapack.dylib; LAPACK version 3.11.0

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

time zone: Europe/Paris
tzcode source: internal

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

other attached packages:
[1] flowCore_2.14.2

loaded via a namespace (and not attached):
[1] compiler_4.3.3 RProtoBufLib_2.14.1 cytolib_2.14.1 Biobase_2.62.0 S4Vectors_0.40.2 BiocGenerics_0.48.1 matrixStats_1.3.0 stats4_4.3.3

sessionInfo() in Windows:

R version 4.4.0 (2024-04-24 ucrt) Platform: x86_64-w64-mingw32/x64 Running under: Windows 10 x64 (build 17763)

Matrix products: default

locale:
[1] LC_COLLATE=English_United States.1252
[2] LC_CTYPE=English_United States.1252
[3] LC_MONETARY=English_United States.1252
[4] LC_NUMERIC=C
[5] LC_TIME=English_United States.1252

time zone: Europe/Paris
tzcode source: internal

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

other attached packages:
[1] flowCore_2.16.0

loaded via a namespace (and not attached):
[1] compiler_4.4.0 RProtoBufLib_2.16.0 cytolib_2.16.0
[4] Biobase_2.64.0 S4Vectors_0.42.0 BiocGenerics_0.50.0
[7] matrixStats_1.3.0 stats4_4.4.0

[SamGG]

Hi,
In fact, the header of Sample_A is wrong (explanations below).
FlowCore could take this error into account, but this will not correct a bug issued by the instrument. @mikejiang should/will you?
Meanwhile, if you have only a few files to correct, I suggest to use HxD 64bits with caution.

My best wishes to Bernd.
Best,
Samuel

In B, the end of the text is at 4902, and the start of data is 4903. Both are on this screenshot.

These values are correct and agree with the data in the screenshot below.

In A, the end is incorrect (no idea why)

The real end is at 4888 as in the following screenshot.

Use HxD and change 6768 into 4888 and 6769 into 4889, and magic might appear... but as the end of data is also wrongly reported, you should have a little more work if flowCore doesn't manage it.

[gfinak]

Thanks, @SamGG.
My 2c here, @CroixJeremy2, your instrument generated an invalid file ( in that it set the begin and end of data / text wrong in the header). You can certainly manually fix this...but why would you trust the data from this file.. who knows what else the software / instrument corrupted. If this data matters in any way.. I wouldn't trust it.