Ch3MS-RF: a random forest model for chemical characterization and improved quantification of unidentified atmospheric organics detected by chromatography–mass spectrometry techniques

Abstract

Abstract. The chemical composition of ambient organic aerosols plays a critical role in driving their climate and health-relevant properties and holds important clues to the sources and formation mechanisms of secondary aerosol material. In most ambient atmospheric environments, this composition remains incompletely characterized, with the number of identifiable species consistently outnumbered by those that have no mass spectral matches in the literature or the National Institute of Standards and Technology/National Institutes of Health/Environmental Protection Agency (NIST/NIH/EPA) mass spectral databases, making them nearly impossible to definitively identify. This creates significant challenges in utilizing the full analytical capabilities of techniques which separate and generate spectra for complex environmental samples. In this work, we develop the use of machine learning techniques to quantify and characterize novel, or unidentifiable, organic material. This work introduces Ch3MS-RF (Chemical Characterization by Chromatography–Mass Spectrometry Random Forest Modeling), an open-source, R-based software tool, for efficient machine-learning-enabled characterization of compounds separated in chromatography–mass spectrometry applications but not identifiable by comparison to mass spectral databases. A random forest model is trained and tested on a known 130 component representative external standard to predict the response factors of novel environmental organics based on position in volatility–polarity space and mass spectrum, enabling the reproducible, efficient, and optimized quantification of novel environmental species. Quantification accuracy on a reserved 20 % test set randomly split from the external standard compound list indicates that random forest modeling significantly outperforms the commonly used methods in both precision and accuracy, with a median response factor percent error of −2 %, for modeled response factors, compared to > 15 %, for typically used proxy assignment-based methods. Chemical properties modeling, evaluated on the same reserved 20 % test set and an extrapolation set of species identified in ambient organic aerosol samples collected in the Amazon rainforest, also demonstrate robust performance. Extrapolation set property prediction mean absolute errors for carbon number, oxygen to carbon ratio (O : C), average carbon oxidation state (OSc‾), and vapor pressure are 1.8, 0.15, 0.25, and 1.0 (log(atm)), respectively. Extrapolation set out-of-sample R2 for all properties modeled are above 0.75, with the exception of vapor pressure. While predictive performance for vapor pressure is less robust compared to the other chemical properties modeled, random-forest-based modeling was significantly more accurate than other commonly used methods of vapor pressure prediction, decreasing the mean vapor pressure prediction error to 0.24 (log(atm)) from 0.55 (log(atm)) (chromatography-based vapor pressure prediction) and 1.2 (log(atm)) (chemical formula-based vapor pressure prediction). The random forest model significantly advances an untargeted analysis of the full scope of chemical speciation yielded by two-dimensional gas chromatography (GCxGC-MS) techniques and can be applied to gas chromatography coupled with electron ionization mass spectrometry (GC-MS) as well. It enables the accurate estimation of key chemical properties commonly utilized in the atmospheric chemistry community, which may be used to more efficiently identify important tracers for further individual analysis and to characterize compound populations uniquely formed under specific ambient conditions.

Anthropogenic and Biogenic Contributions to the Organic Composition of Coastal Submicron Sea Spray Aerosol

“…A representative external standard containing ∼130 known components was analyzed in 6-point calibration curves immediately preceding and following sample analysis. Compounds present in both the sample and standard or with clear chemical proxies in the standard were directly quantified by the calibration curves of the external standard, while the quantification factors of all other species were predicted using Ch3MS-RF . Model performance and uncertainties of both modeled and directly calculated quantification factors are described and illustrated in Supporting Information SI.2.…”

“…Compound populations exhibiting more complicated temporal variability or containing compounds with conflicting source groups described in the literature were defined as mixed influence. Finally, the chemical properties of unidentifiable compounds, specifically average carbon oxidation state ( true normalO normalS normalc 0em ̅ ) and carbon number ( n c ), were predicted using Ch3MS-RF, as described in detail in Supporting Information SI.7. Visualizations of complex organic mixtures in the n c – true normalO normalS normalc 0em ̅ space are commonly utilized in organic aerosol analysis. − As both metrics are calculable from the chemical formula, orientation of GC × GC–MS results in n c – true normalO normalS normalc 0em ̅ space facilitates comparison with analyses by techniques that identify compounds by the chemical formula such as high-resolution chemical ionization mass spectrometry.…”

“…Using the custom mass spectral library, all ∼750 catalogued species were traced over the entire bloom. Each observation of each compound was normalized by the mean of the three nearest internal standard components to correct for drifts in instrument sensitivity over the analysis period and matrix effects using the methodology described in detail in Franklin et al, 2022 . Finally, the mass spectrum of each compound was searched against the NIST-14 mass spectral database.…”

“…Each observation of each compound was normalized by the mean of the three nearest internal standard components to correct for drifts in instrument sensitivity over the analysis period and matrix effects using the methodology described in detail in Franklin et al, 2022. 34 Finally, the mass spectrum of each compound was searched against the NIST-14 mass spectral database. Of the ∼750 compounds catalogued, 14% were identifiable by database match.…”

“…Compound populations exhibiting more complicated temporal variability or containing compounds with conflicting source groups described in the literature were defined as mixed influence. Finally, the chemical properties of unidentifiable compounds, specifically average carbon oxidation state (OS c ) and carbon number (n c ), were predicted using Ch3MS-RF, 34 as described in detail in Supporting Information SI.7. Visualizations of complex organic mixtures in the n c −OS c space are commonly utilized in organic aerosol analysis.…”