73rd Conference on Mass Spectrometry and Allied Topics

Anne Marie Smith¹; Alexander Bourque¹; Artyom Petrovskiy¹; Alexander Sakharov¹; Eduard Kolovanov¹; Vitaly Lashin¹

¹ACD/Labs, Toronto, ON

Introduction

The food, fragrance, and flavor industries depend on a delicate balance of components to achieve their signature profiles. To enhance product quality, drive innovation, and maintain competitiveness, precise comparisons of these complex mixtures are essential. Effective quality control ensures consistency, while flavor and aroma profiling help refine formulations and identify key compounds. Evaluating ingredient sourcing and authenticity is crucial for upholding quality standards, while understanding that shelf life and regulatory compliance fosters consumer trust. However, traditional analytical methods are labor-intensive and prone to error. Here, we present a streamlined solution for GC-MS data analysis, databasing, and reporting, enabling straightforward comparisons and efficient exportation of results for deeper analysis.

Methods

We employed a componentization algorithm to separate co-eluting components through peak detection and ion grouping. The algorithm also includes spectral searching capabilities of specified databases. The identified components are uploaded to a searchable database, facilitating efficient data management. Samples are selected for comparison, and comparative reports are generated, and exported in a tabular format for further analysis.

Preliminary Data

To demonstrate the benefits of componentization and comparative analysis, we analyzed four GC-MS fragrance datasets sourced from open-data repositories. Optimization settings were established to extract the maximum number of likely real components from these datasets. Following this, spectral searches were conducted using the NIST/EPA/NIH Mass Spectral Library (2023 edition), with the libraries converted to ACD/Labs database format. The hits returned included structures and retention indices from the commercial library. All identified components were accurately updated in a local database.

The comparison was executed by a single click to analyze the four datasets, revealing both common and unique components across the fragrances. The generated report includes critical data such as CAS numbers, retention times, and areas, enabling users to identify key compounds and their relative amounts. This capability is essential for various applications, including the development of unique formulations and the maintenance of quality control, ensuring consistency across production batches. Furthermore, understanding the variations between datasets can aid in testing authenticity and market analysis.

Our tool significantly enhances the efficiency of identifying unknown components in complex mixtures. The resultant table organizes parameters such as name, CAS number, area, and retention time by dataset, facilitating quick discernment of differences. This organization supports informed decision-making, enabling stakeholders to leverage data effectively in product development and quality assurance processes.

Novel Aspect

Our approach integrates advanced componentization with user-friendly databasing, allowing efficient comparisons driving innovation across the food, fragrance, and flavor industries.

Alexander Bourque¹; Anne Marie Smith¹; Artyom Petrovskiy¹; Vitaly Lashin¹

¹ACD/Labs, Toronto, ON

Introduction

Peptides are integral to numerous biological processes and have already been applied in medicine, functioning as hormones, neurotransmitters, and signaling molecules. Their selective binding to receptors makes them invaluable for drug development, particularly in targeted therapies. They also serve as biomarkers for disease diagnosis and progression. Research into synthetic peptides is expanding, paving the way for novel therapeutics and enhanced vaccine efficacy, positioning them as a focal point in biochemistry and pharmacology. Here, we introduce a robust data analysis approach for LC-MS spectra of peptide compounds. Our method incorporates in silico fragmentation, including new rules for generating multiply charged species and fragmenting amide and disulfide bonds, with clear spectral highlighting and identification of structures and ion types.

Methods

MSn spectra of peptide samples were analyzed using various mass accuracies. Structures were assigned to spectra based on optimized in silico fragmentation rules, aiming for the highest assignment scores. The mass differences between the experimental spectra and theoretical fragment formulas were evaluated to assess assignment precision. To evaluate the impact of these enhancements on overall results, comparisons were made between the previously available settings and the newly developed peptide-specific settings for multiply charged species, as well as for amide and disulfide bond fragmentation.

Preliminary Data

In silico fragmentation was applied to select LC-MS and extracted MSn spectra. Preliminary data indicated improved assignment scores due to the introduction of new fragmentation rules. The visualization of fragments on the spectra, along with the provided mass differences, facilitated the evaluation of assignment accuracy. This method effectively managed both multiply charged parent species and fragment ions. Additionally, it was successfully applied to various isomeric structures within single mass spectra in a semi-automated manner, enhancing our ability to distinguish closely related isomers. Fragmentation schemes, along with spectra annotated with labeled a, b, c, x, y, and z ions, were readily generated from the configurable reporting engine. This capability not only streamlines complex peptide analysis but also allows scientists to focus more on their research and less on the intricacies of fragmentation rules, ultimately enhancing peptide data analysis.

Novel Aspect

Integration of peptide fragmentation rules with automated labeling of a, b, c, x, y, and z ions for enhanced analysis.