Re: [proteowizard-support] Thermo RAW file corruption checker (was: Error during conversion for a large HCD file from Orbi Velo)

[Matthew C. <mat...@gm...>]

Hi Wen,

There's no easy way to handle this. Once the API queries those scans, the process is potentially 
compromised and you can't trust the data coming out. Trying to continue after those scans would 
cause a crash anyway, just like QualBrowser. If we knew a priori which scans were bad, we could 
exclude them by their index.

Since this isn't the first time it's been an issue, I went ahead and created a standalone program 
that will check a RAW file and report corrupt scans. I also made it report the inverse set of scan 
intervals (the non-corrupt scans) which are suitable for feeding to msconvert like:
--filter "scanNumber <2nd line of output from CorruptionChecker>"

CorruptionChecker starts itself over (at the next scan) after any failed operation. However, while 
testing whether it worked, I found that simply excluding the apparently corrupt scans from the 
scanNumber set still caused crashes. So I tried excluding the adjacent scans (-1 and +1) too and it 
still crashed. So I went to -2 and +2 and it stopped crashing. I guess that's better than nothing. 
For other files -2/+2 might be too much or not enough; I only had one corrupt file to test on.

Hope this helps,
-Matt


On 10/25/2012 2:42 PM, Yu, Wen wrote:
> By the way, multiple files from this dataset have the same problem, though at different point of the runs. Strange!
>
> Wen
>
>
> -----Original Message-----
> From: Matthew Chambers [mailto:mat...@gm...]
> Sent: Thursday, October 25, 2012 12:36 PM
> To: Yu, Wen
> Cc: su...@pr...
> Subject: Re: [proteowizard-support] Error during conversion for a large HCD file from Orbi Velo
>
> Looks like data corruption. Can you view that scan in QualBrowser (both the peaks and the scan header)?
>
> -Matt
>
>
> On 10/25/2012 2:26 PM, Yu, Wen wrote:
>> Hi, we're see a consistent error as detailed below. The test file is a ~1.3GB HCD run from Mann's
>> lab. The same error occurs with the most recent release of msconvert as well as the version from
>> Nov, 2011.
>>
>> Hope you can shed some light on the cause and solution to this problem. I'll be happy to send you
>> the raw file if have to.
>>
>> Thanks!
>>
>> Wen Yu
>>
>
>> [SpectrumList_Thermo::spectrum()] Error retrieving spectrum "controllerType=0 controllerNumber=1
>> scan=1333": [ThermoRawFile] [RawFileImpl::getMassList(), GetMassListFromScanNum()] Failed call to XRawfile: The system
>> cannot find the file specified.