SWATH-MS Data Reproducibility Study

In recent years, proteomics research methods have been increasingly widely applied, with various proteomics research approaches continually emerging, enhancing detection accuracy, sensitivity, and breadth. The starting point for proteomics research is often to study the expression characteristics of multiple proteins on a large scale to identify new biomarkers. As a result, shotgun proteomics, as a method for large-scale analysis, was very suitable in the early stages of proteomics research. However, although shotgun proteomics is broad in scope, its identification accuracy is relatively low. In contrast, targeted proteomics excels in precision, but the number of proteins identified may be reduced. Therefore, researchers have been seeking a proteomics research method that combines both precision and breadth. SWATH fulfills these two characteristics.

This issue shares a study on the reproducibility of SWATH across different laboratories, published in the journal Nature Communications this April.

Study on the Reproducibility of SWATH-MS Data - Bruker BioSpin

SWATH is truly a panoramic mass spectrometry detection method capable of detecting nearly all molecules in complex samples. The principle of SWATH is to divide the MS spectrum into multiple small windows, then perform MS/MS detection on each ion in each window, thereby achieving deep detection of all molecules. The detection precision of SWATH is comparable to that of MRM targeted proteomics. Combining the advantages of both breadth and precision, SWATH is an ideal method for researchers studying proteomics.

However, prior to this, the reproducibility of SWATH research data had only been well-verified within individual laboratories, and its reproducibility across different laboratories had not been validated. Due to the numerous influencing factors often affecting complex samples, the reproducibility of data is relatively low. This article primarily investigates the issue of data identification reproducibility of SWATH between different laboratories around the world. The study collected data on reproducibility from 11 different laboratories worldwide using SWATH.

Experimental Results

1. The authors divided 30 peptide standards into 5 groups, with 6 standards per group, mixing them into HEK293T cell lysates at 5 dilution gradients. These samples were then analyzed using the SWATH method in 11 laboratories across 6 selected countries. A total of 229 SWATH-MS files were obtained, and subsequent analysis was conducted on data such as the linear dynamic range.

2. The results indicated that the total number of proteins identified across the 11 laboratories was generally consistent, typically around the range of 4000.

3. The authors further analyzed the range of coefficients of variation within the same laboratory and between different laboratories, showing that the coefficients of variation for both the standards and HEK293 proteins were approximately 20%.

4. Additionally, the dynamic range analysis of the standards and samples also presented good linear results.

5. The authors conducted clustering and correlation analysis on the proteins identified by the 11 laboratories, and the data exhibited high symmetry, indicating that SWATH-MS identification and data analysis can achieve high reproducibility across different laboratories.

Conclusion
In summary, the results indicate that SWATH can repeatedly identify at least 4000 proteins in HEK293T cells across these 11 laboratories. Furthermore, the coefficient of variation for protein identification is low, the proteins identified display similar linear dynamics, and clustering and correlation analyses are consistent. Therefore, the conclusion is that SWATH not only has high reproducibility within individual laboratories but also exhibits extremely high reproducibility across laboratories, further validating the reliability of SWATH identification.