Re: [Open-ms-general] openms preprocessing tools

[Chris B. <bi...@pc...>]

Hey Tony,

the command line actually does allow this.
You can split the filename in the for-in-do loop into its components (drive 
letter, path, filename-prefix and filename-suffix) and insert whatever you want 
in between. See example below where I inserted "_filtered" just before the 
suffix (this would convert "hello.txt" to "hello_filtered.txt")


for  %%I  in (*.*) do @echo FileFilter -in %%I -out %%~dI%%~pI%%~nI_filtered%%~xI

REM explanation:
REM "%%~dI%%~pI%%~nI" gives the complete filename except the suffix
REM "%%~xI"           gives the suffix of %%I
REM see http://technet.microsoft.com/de-de/library/bb490909(en-us).aspx
REM for details


A more basic approach is simply to reuse the whole filename and just prepend or 
append something, i.e.

REM basic version
for  %%I  in (*.*) do @echo FileFilter -in %%I -out filtered_%%I

You will obviously need to adapt the regular expression (*.*) and the actual 
command (everything after @echo) to your needs :)

hope this helps.

cheers
Chris

Larson, TR wrote:
> Dear Chris,
> Thanks for the quick reply - that was exactly the information I needed. 
>   I will run some tests.
> 
> I have one more question.  Is there an easy way to batch signal 
> processing for several files?  I'm talking hundreds of files here, and I 
> don't want tot  do them one by one.  I have run batches using Windows 
> command line with reAdW for converting thermo .raw to .mzXML format 
> using for-in-do syntax. This works because reAdW only needs the source 
> file and converts to the output format with no other user input needed. 
>   I assume this won't work for the openms tools because an -out filename 
> needs to be explicitly specified for each file?  Is there a way to 
> automatically specify an output name to a file while batching - e.g. to 
> append  "filtered" to the input file "test.mzData" so that it is 
> automatically renamed to "test_filtered.mzData" during a batch execution 
> of baselineFilter?
> 
> 
> thanks
> Tony
> 
> 
> 
> 
> thnaks
> Tony
> 
> - however, this
> 
> Chris Bielow wrote:
>> Hi Tony,
>>
>> Larson, TR wrote:
>>> Hello,
>>> I am completely new to using openms; yesterday I downloaded openms 1.2 
>>> to have a play around with it.  My primary interest is in processing and 
>>> peak-picking from LCMS and GCMS .cdf data files. I was initially pointed 
>>> to openms from users of the R package, xcms, because I needed to find a 
>>> tool to pre-process my .cdf files; specifically to remove noise from the 
>>> chromatograms before further processing.
>>>
>>> openms is the only package I found that can do this with .cdf files. 
>>> I've managed to work out how to use BaselineFilter, FileFilter, and 
>>> FileConverter from the TOPP command line interface.
>>>
>>> What I am typically doing is using FileConverter to get my .cdf files 
>>> into .mzData format, then applying BaselineFilter to reduce the baseline 
>>> noise, and finally using FileFilter to remove the residual noise below 
>>> an intensity threshold.
>>>
>>> However, I'm confused because  the Signal Processing tutorial in the 
>>> openms .html documentation lists gaussian and savitzy-golay smoothing 
>>> filters in addition to the baseline filter - but I can't find either of 
>>> these from the TOPP command line.  
>> Have a look at NoiseFilter.exe
>> which has a -type parameter:
>>
>> Options (mandatory options marked with '*'):
>>    -in <file>*          Input raw data file  (valid formats: 'mzData')
>>    -out <file>*         Output raw data file  (valid formats: 'mzData')
>>    -type <type>*        Smoothing filter type (valid: 'sgolay,gaussian')
>>
>>
>>
>>> I'm also interested in other signal 
>>> noise processing, such as per-spectra thresholding, but there doesn't 
>>> seem to be a tool in openms for this.
>> You could for example use FileFilter.exe with its -sn <number> Option.
>> <number> is the minimal Signal-To-Noise ratio a peak needs to have in order to 
>> pass the filter. It works scanwise.
>> Use '1' to filter about 50% of the data (as the noise is estimated using the 
>> median of a sliding window). Higher numbers will filter out more peaks (with 
>> e.g. '10' being very aggressive), lower SN values will filter out fewer peaks. 
>> Play with it a little until you find the settings which fit your needs.
>>
>>> I'm not a programmer, but I do want to make best use of openms.  Can 
>>> someone please help me to understand how to apply these other signal 
>>> processing techniques to my data?  Is this possible from the TOPP 
>>> command line using one of the .exe files or must I do it another way? 
>>> I've looked through all the documentation and I can't work it out.
>> I hope the above will get you started. There is also the possibility to resample 
>> your data using Resampler.exe to make the peaks equidistant, but that is only 
>> needed in special cases.
>>
>> If you have more questions, don´t hesitate to ask :)
>>
>> cheers
>> Chris
>>
>>
>>> thanks
>>> Tony
>>>
>>>
>>> -------------------------------------------------------------------------
>>> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
>>> Build the coolest Linux based applications with Moblin SDK & win great prizes
>>> Grand prize is a trip for two to an Open Source event anywhere in the world
>>> http://moblin-contest.org/redirect.php?banner_id=100&url=/
>>> _______________________________________________
>>> Open-ms-general mailing list
>>> Ope...@li...
>>> https://lists.sourceforge.net/lists/listinfo/open-ms-general
> 

-- 
Freie Universitaet Berlin
Institut fuer Informatik              Phone: +49 30 838 75244
AG Algorithmische Bioinformatik       room 011
Takustr. 9
D-14195 Berlin