Related Experiment Video
Updated: Jun 15, 2026
13:37
Polymer Microarrays for High Throughput Discovery of Biomaterials
Published on: January 25, 2012
14.9K
ToF-SIMS and Machine Learning for Single-Pixel Molecular Discrimination of an Acrylate Polymer Microarray
Wil Gardner1,2,3, Andrew L Hook4, Morgan R Alexander4
1Centre for Materials and Surface Science and Department of Chemistry and Physics, La Trobe University, Melbourne, Victoria 3086, Australia.
Analytical Chemistry
|April 3, 2020
Summary
Machine learning with self-organizing maps (SOM) successfully analyzed polymer microarrays using time-of-flight secondary ion mass spectrometry (ToF-SIMS) data. This method precisely identified 70 unique polymers and detected variations within spots, advancing materials discovery.
Area of Science:
- Materials Science, Polymer Chemistry, Analytical Chemistry, Data Science
Background:
- Combinatorial approaches and polymer microarrays accelerate novel polymer development by enabling high-throughput property comparison.
- Accurate discrimination of similar polymer chemistries and detection of heterogeneities within polymer spots are critical challenges.
- Time-of-flight secondary ion mass spectrometry (ToF-SIMS) provides high-resolution chemical information from sample surfaces but generates complex hyperspectral data.
Purpose of the Study:
- To apply machine learning, specifically self-organizing maps (SOM), to analyze ToF-SIMS data from polymer microarrays for the first time.
- To demonstrate the capability of SOM in achieving single-pixel molecular discrimination of numerous unique polymer spots.
- To identify and characterize intraspot heterogeneities and relate polymer chemistry to performance.
Main Methods:
- Utilized a previously established machine learning method involving color-tagging the output of a self-organizing map (SOM) to interpret ToF-SIMS hyperspectral images.
- Applied this methodology to a ToF-SIMS image of a printed polymer microarray containing 70 unique homopolymer spots.
- Correlated the SOM analysis results with fluorescence data from polymer-protein adsorption studies.
Main Results:
- Achieved complete, single-pixel molecular discrimination of all 70 unique homopolymer spots on the microarray.
- Identified intraspot heterogeneities, likely due to intermixing between the polymer and the pHEMA coating.
- Demonstrated that SOM can cluster data based on polymer backbone structures and side groups, revealing layers of similarity.
Conclusions:
- Self-organizing maps provide a powerful tool for interpreting complex ToF-SIMS data from polymer microarrays, enabling precise molecular discrimination.
- The SOM methodology successfully identifies both distinct polymer spots and subtle intraspot variations, crucial for understanding material properties.
- This approach facilitates the visualization of polymer performance within the broader data landscape, advancing materials discovery and development.
