COMPASS Technology
The COMPASS Analytical Framework
Modern biological instruments can produce far more information than can practically be interpreted one measurement at a time.
COMPASS was developed to work with biological data at this scale—organizing large measurement spaces, comparing them across sample populations, and identifying relationships that help focus subsequent molecular and biological investigation.
For mass spectrometry in particular, this creates the opportunity to combine greater analytical depth within each sample with comparisons across larger sample populations, while also relating mass spectrometry measurements to information generated by other omic technologies.
Expanding the Analytical Scale of Mass Spectrometry
Modern mass spectrometry can generate hundreds of thousands to millions of measurable datapoints from a complex biological sample.
As the number of samples increases, the analytical problem grows rapidly. Conventional workflows often manage that complexity by reducing the data before comparative analysis begins—through targeted acquisition, feature selection, filtering, or other forms of data reduction.
Those approaches can be useful, but they also narrow the measurement space available for discovery.
COMPASS takes a broader approach. It is designed to retain and comparatively analyze large portions of the underlying measurement space across many samples, allowing the data itself to indicate which regions contain information associated with the biological question being studied.
This changes an important constraint on mass spectrometry research.
Greater data depth per sample can be combined with larger sample populations, allowing mass spectrometry to be used not only for identifying and measuring selected molecules, but also for discovering patterns across complex biological populations.
From Measurement Space to Biological Signal
COMPASS approaches a complex dataset as a population of comparable measurements rather than simply a list of previously identified molecules.
The analytical process begins with the underlying measurement space. Corresponding regions can be compared across samples and evaluated for their relationship to a defined biological classification, experimental condition, or outcome.
Measurements showing stronger relationships can then be prioritized for deeper investigation.
This allows discovery to proceed without requiring the molecular identity of every potentially informative signal to be established beforehand.
For mass spectrometry data, the analysis can draw on characteristics such as signal abundance, retention time, mass-to-charge relationships, and patterns observed across sample populations. Molecular characterization and MS2 information can then add additional context to signals identified as informative.
The result is a progression from large-scale measurement → comparative analysis → prioritized signals → molecular and biological interpretation.
Connecting MS1, MS2, and Multi-Omic Data
No single analytical technology provides a complete description of a biological system.
MS1 measurements provide broad quantitative information about molecular signals across samples. MS2 fragmentation data can contribute additional information for molecular characterization and identification. Genomic, transcriptomic, proteomic, metabolomic, and other measurements provide additional views of the same underlying biology.
COMPASS provides a common computational context in which these different forms of information can be related.
Rather than treating each analytical layer as an independent endpoint, measurements can be associated across samples and data types so that relationships discovered in one layer can be examined alongside information from others.
This creates the possibility of linking mass spectrometry signals with complementary molecular measurements and investigating biological patterns that may be difficult to recognize when each dataset is considered separately.
The combination of analytical depth, population-scale comparison, and integration across data types substantially broadens the kinds of biological questions that can be explored with mass spectrometry.
Research Made Possible by a Broader Analytical View
The COMPASS approach can be applied wherever meaningful biological information may be distributed across large numbers of measurements rather than concentrated in a small number of predetermined targets.
Biomarker and Molecular Signature Discovery
Identify individual signals or combinations of measurements associated with biological states, disease characteristics, or experimental outcomes.
Treatment Response
Compare populations or paired samples to identify molecular changes associated with therapeutic response and other treatment-related effects.
Disease and Population Characterization
Examine patterns distributed across many measurements to characterize differences among biological states, cohorts, or sample populations.
Multi-Omic Relationships
Relate mass spectrometry measurements to complementary molecular data to investigate associations across different layers of biology.
Molecular and Pathway Investigation
Use computationally prioritized signals to focus subsequent molecular identification and connect analytical findings with biological pathways and mechanisms.
Hypothesis Generation
Allow unexpected relationships within the experimental data to point toward biological questions that were not defined before the analysis began.
COMPASS in Practice
The analytical approach underlying COMPASS has been evaluated using real biological datasets ranging from blinded treatment-response samples to peer-reviewed studies of disease progression.
These studies illustrate how broad mass spectrometry measurement spaces can be compared across biological populations, reduced computationally to informative regions, and subsequently connected with molecular and biological interpretation.