Annotator: Postprocessing Software for Generating Function-based Signatures from Quantitative Mass Spectrometry

Mass spectrometry is used to investigate global changes
in protein
abundance in cell lysates. Increasingly powerful methods of data collection
have emerged over the past decade, but this has left researchers with
the task of sifting through mountains of data for biologically significant
results. Often, the end result is a list of proteins with no obvious
quantitative relationships to define the larger context of changes
in cell behavior. Researchers are often forced to perform a manual
analysis from this list or to fall back on a range of disparate tools,
which can hinder the communication of results and their reproducibility.
To address these methodological problems, we developed Annotator,
an application that filters validated mass spectrometry data and applies
a battery of standardized heuristic and statistical tests to determine
significance. To address systems-level interpretations, we incorporated
UniProt and Gene Ontology keywords as statistical units of analysis,
yielding quantitative information about changes in abundance for an
entire functional category. This provides a consistent and quantitative
method for formulating conclusions about cellular behavior, independent
of network models or standard enrichment analyses. Annotator allows
for “bottom-up” annotations that are based on experimental
data and not inferred by comparison to external or hypothetical models.
Annotator was developed as an independent postprocessing platform
that runs on all common operating systems, thereby providing a useful
tool for establishing the inherently dynamic nature of functional
annotations, which depend on results from ongoing proteomic experiments.
Annotator is available for download at http://people.cs.uchicago.edu/∼tyler/annotator/annotator_desktop_0.1.tar.gz.