Let’s revisit the DNR geoduck mass spec data!
After the initial DNR geoduck data processing (MSConvert, Pecan, Skyline), we’ve learned that there are a few changes that need to be made. Specifically:
- When converting .raw files (from Lumos MSMS) to .mzml, I need to demultiplex the data. What is demultiplexing you ask? More on that later.
- When running Pecan, do all my sample files first (including my blanks). Then, go back and run the “cleaning” blank files after.
- Need to use Genome Sciences server, Hyak, or some other powerful machine to run Pecan, since our computers don’t have enough memory
Demultiplexing .raw files and converting to .mzML
NOTE: all the programs needed for processing Lumos data files require Windows. For this workflow I’m using the Roberts Lab’s Woodpecker machine
Step 1: Download .raw files from Owl to local folder on Windows computer.
Step 2: Extract the files from the .zip folder didn’t work, as only a few files extracted correctly and the rest resulted in 0kb files. I thus downloaded WinZip program to computer and unzipped.
Step 3: Open command window. One Windows search for: cmd.exe
Step 4: Change working directory to the folder containin .raw files:
cd C:\Users\srlab\Documents\geo **Step 5:** Use the following script outlined in the [DIA data Analyses](http://ift.tt/2lGfTdS) instructions, with a couple directory edits. This first run only includes 1 file, so I can see if I get any errors: C:\Users\srlab\Documents\oystertest\20170412_pwiz_testBuild_addMinWindowSize\msconvert.exe –zlib –64 –mzML –filter “peakPicking true 1-2” –filter “demultiplex optimization=overlap_only” 2017_January_23_envtstress_geoduck1.raw`
Step 6: No errors! To run more files in tandem, open up more terminal windows, navigate to the directory housing the .raw files, and enter the same script (with different file names). I have 4 windows running at the same time, on geoduck1-geoduck4.
Step 7: Wait! Started on Friday 3/14 @ 2pm. Let’s see when they finish.