Entering edit mode
lloyd.izard
▴
80
@lloydizard-15069
Last seen 5.9 years ago
Good morning,
I am trying to read from a FCS file but I get the following error:
Error in readFCSdata(con, offsets, txt, transformation, which.lines, scale, :
$PnRNAis larger than R's numeric limit:1.79769313486232e+308
In addition: Warning message:
In readFCSdata(con, offsets, txt, transformation, which.lines, scale, :
NAs introduced by coercion
The thing is, when I try to read the same file in .csv format, I can read it!
# library
library('flowCore')
library('Biobase')
library('data.table')
# data
data <- read.table('file.csv', dec = ".", sep = ",")
data <- read.FCS('file.fcs', alter.names = F, transformation = "linearize")
My question is: is there a memory limit for read.FCS()? And if so, how do I extend it?
All the best, Lloyd Izard
Hi,
I don't suspect a memory limit there. The read.FCS has been extended recently to cope with file bigger than 2GB. The error sounds like a problem in reading and interpreting the TEXT segment of the FCS file. Of course reading a CSV file avoids this problem because there is no such meta data. I think you should report the error onto https://github.com/RGLab/flowCore. You should report the sessionInfo() result and share some information about the FCS: instrument...
Meanwhile, you could verify that the FCS complies with FCS standard using flowIO http://bioinformin.cesnet.cz/flowIO/ (CLIP, Prague) or https://primitybio.github.io/fcs-validator/ (Zach Bjornson).
Best.