Quantification algorithms for multiplexed data-independent acquisition data in shotgun proteomics

Alexeev, Dmitry (2026): Quantification algorithms for multiplexed data-independent acquisition data in shotgun proteomics. Dissertation, LMU München: Fakultät für Chemie und Pharmazie

Abstract

Proteomics is a rapidly growing research field. Recent advancements towards data-independent acquisition (DIA) are overcoming limitations of the data-dependent approaches (DDA), such as the limited amount of analyzed ions per run and the exclusion of fragmentation signals from the quantification process. Nevertheless, techniques developed at first to boost the DDA performance can be beneficial to the DIA workflow. As such, multiplexing techniques found their way into the newly emerging field of DIA proteomics. The concept of multiplexing implies the use of labels – be it isotopically labeled amino acids or chemical groups attached to the amino acids during sample preparation. By marking proteins from each sample via its unique label, researchers can analyze them simultaneously in a single run. While improving the throughput of the experiment, multiplexing can also boost the identification rate by transferring identifications between channels. This study’s goal is to provide a computational platform based on the established MaxDIA workflow to analyze multiplexed DIA proteomics samples termed MultiplexDIA. Moreover, the existing MaxLFQ algorithm was generalized to account for multiplexed signals for quantification. The results of this study highlight the algorithm’s structure and its performance against the DIA-NN software on datasets with normal and single-cell-like amounts of proteomes in the samples. Different types of labels, as well as different modes of identification transfer between labels, were tested. Our results suggest that MultiplexDIA is able to achieve similar levels of improvement as DIA-NN in identification rate, quantification accuracy, and false discovery rate (FDR) control when comparing results of the multiplexed and label-free analysis. The complete MS1 multiplex transfer proves to be the most resilient mode of identification transfer between labels, allowing for additional identifications without impairment of the quantification. The reworked MaxLFQ can normalize and quantify each multiplexed channel separately, thus retaining more capabilities for throughput. Both metabolic and chemical non-isobaric labels are suitable for the MultiplexDIA workflow, though the former show better performance.

• Dokumententyp: Dissertationen (Dissertation, LMU München)
• Keywords: Proteomics, multiplexing, computational biology, quantification, single-cell
• Themengebiete: 500 Naturwissenschaften und Mathematik ; 500 Naturwissenschaften und Mathematik > 540 Chemie
• Fakultäten: Fakultät für Chemie und Pharmazie
• Sprache der Hochschulschrift: Englisch
• Datum der mündlichen Prüfung: 16. Januar 2026
• 1. Bericht­erstatter:in: Klughammer, Johanna