R2DGC

Please note that this code is still slow and doesn't quite do everything yet. I'll send updates as I improve on it. I'll be performing the following steps:

• Step 1: fully implement all functionality.

• Step 2: change up the data types to use exclusively numpy arrays.

• Step 3: parallelise the code.

To run any of the Python code, you will need to import the module, e.g.:

import R2DGC as r

Put that at the top of your script. You should also import pandas and numpy. See the QuickStart.py file for an example.

In your script you can then call each function directly from the module, e.g.:

Alignment = r.ConsensusAlign(inputFileList)

Then to run the script, type:

$ python <scriptname.py>

Another note: this code assumes that there is a directory with some files, a directory on the same level called 'processed', and another called 'FAME'. See the Quickstart.py file for an example.

FindProblemIons

This code should work exactly the same as the R code, with the exception of plotting. It takes the following arguments:

inputFile str possibleIons Series, default pd.Series(np.arange(70,601,1)) absentIonThreshold float, default 0.01 commonIonThreshold int, default 2

Returns a list of problem ions.

PrecompressFiles

Also works the same as the R code. This should work for both TIC and UM data. It takes the following arguments:

inputFileList list of strs RT1Penalty int, default 1 RT2Penalty int, default 1 similarityCutoff int, default 95 commonIons list, default [] quantMethod str, default 'T' outputFiles bool, default False

Note that this assumes that there is a directory called 'processed' one level above the directory with your files. If outputFiles is True, it will write to this directory.

Returns a list of files to be compressed.

Find_FAME_STANDARDS

For this to work you need the FIND_FAME_FRAME.txt to be one level above your files by default. You can also supply the path to this file. It takes the following arguments:

inputFileList list of strs FAME_Frame str, default "../FIND_FAME_FRAME.txt" RT1Penalty int, default 1 RT2Penalty int, default 10 similarityCutoffWarningThreshold int, default 80

This also assumes that there is a directory called 'FAME' one level above the directory with your files. It will write to this directory.

ConsensusAlign

Partially implemented, for TIC files only at the moment. Also does not handle multiple seed files and currently cannot use a standard library. It takes the following arguments:

inputFileList list of strs RT1_Standards list of strs, default None RT2_Standards list of strs, default None seedFile int, default 0 RT1Penalty int, default 1 RT2Penalty int, default 10 autoTuneMatchStringency bool, default True, similarityCutoff int, default 90, disimilarityCutoff int, default 0, commonIons list, default [] missingValueLimit float, default 0.75 missingPeakFinderSimilarityLax float, default 0.85 quantMethod str, default "T"

Returns an alignment table.

MakeReference

This should work the same way as the R code. It takes the following arguments:

inputFileList list of strs RT1_Standards list of strs, default None RT2_Standards list of strs, default None

Returns a standard library or a list of missing RT indices.

FeihnMatch

Not yet implemented. Coming soon!