Thanks for your fast answer. You're right. The screenshot is from masswolf. Here is the one from R :
Error in writeLines(unlist(log_convert_warning), fileConn) : invalid 'text' argument Then the script stops : close(fileConn) The only created file is a mzXML one.
How would you like me to send you an exemple file ? (approximately 80 Mo) If we cannot read the MSE file, how could I edit the script in order to read only the MS1 trace and the lockmass trace ?
The script worked very well with MS1 data, but I tried it with MSE data (also from a Synapt Waters) and here is the error statement :
First function is not of type MS, is this MS^E data? Continuing anyway. Second function is not of type MS, is this MS^E data? Continuing anyway. (got computer name: C-TAC) Calculating sha1-sum of mzXML --done (mzXML sha1):8160c2c3905aba20aeb020ca843c7e61bd71c8df
Only one file is created and then the script stop at the second one, probably the MSE one ?
I'm currently working on a Waters Synapt (ESI + source) and I encounter an important loss of sensitivity over time, probably due to the source clogging (I'm working with intact proteins). Which normalisation method would you recommend to fix this quasilinear "drift" ? Could we assume that this drift is similar to a dilution effect and correct it using PQN or Median normalisation ?
I encountered the same issue as you Asun, and tried to fix it according to Paul's answer, but I still get the following error message trying to apply the lockMassFreq correction...
Warning in AutoLockMass(xRaw) : Lock mass frequency wasn't detected correctly
Does anyone know how to save a deconvoluted mass spectrum processed by MaxEnt1 (Waters MassLynx deconvolution software) in order to process it on XCMS ?