Feature DetectionΒΆ

One very common task in Mass Spectrometric research is the detection of 2D features in a series of MS1 scans (MS1 feature detection). OpenMS has multiple tools that can achieve these tasks, these tools are called FeatureFinder. Currently the following FeatureFinders are available in OpenMS:

  • FeatureFinderMultiplex
  • FeatureFinderMRM
  • FeatureFinderCentroided
  • FeatureFinderIdentification
  • FeatureFinderIsotopeWavelet
  • FeatureFinderMetabo
  • FeatureFinderSuperHirn

One of the most commonly used FeatureFinders is the FeatureFinderCentroided which works on (high resolution) centroided data. We can use the following code to find Features in MS data:

from pyopenms import *

# Prepare data loading (save memory by only
# loading MS1 spectra into memory)
options = PeakFileOptions()
fh = MzMLFile()

# Load data
input_map = MSExperiment()
fh.load("test.mzML", input_map)

ff = FeatureFinder()

# Run the feature finder
name = "centroided"
features = FeatureMap()
seeds = FeatureMap()
params = FeatureFinder().getParameters(name)
ff.run(name, input_map, features, params, seeds)

fh = FeatureXMLFile()
fh.store("output.featureXML", features)
print("Found", features.size(), "features")

You can get a sample file for analysis directly from here.

With a few lines of Python, we are able to run powerful algorithms available in OpenMS. The resulting data is held in memory (a FeatureMap object) and can be inspected directly using the help(features) comment. It reveals that the object supports iteration (through the __iter__ function) as well as direct access (through the __getitem__ function). We can also inspect the entry for FeatureMap in the pyOpenMS manual and learn about the same functions. This means we write code that uses direct access and iteration in Python as follows:

f0 = features[0]
for f in features:
    print (f.getRT(), f.getMZ())

Each entry in the FeatureMap is a so-called Feature and allows direct access to the m/z and RT value from Python. Again, we can lear this by inspecting help(f) or by consulting the Manual.

Note: the output file that we have written (output.featureXML) is an OpenMS-internal XML format for storing features. You can learn more about file formats in the Reading MS data formats section.