Combining peak- and chromatogram-based retention time alignment algorithms for multiple chromatography-mass...

Area of Science:

• Analytical Chemistry
• Biochemistry
• Computational Biology

Background:

• Modern analytical techniques like GC-MS and LC-MS generate high-dimensional data essential for metabolomics and proteomics.
• Manual comparison of complex datasets with thousands of signals is laborious and prone to errors.
• Accurate alignment and matching of features across multiple experiments are crucial for reliable metabolite or protein identification.

Purpose of the Study:

• To introduce novel algorithms for accurate retention time alignment of multiple GC-MS datasets.
• To address limitations of existing methods that focus on either extracted peaks or raw data alignment.
• To provide robust computational tools for analyzing complex hyphenated separation technique data.

Main Methods:

• Development of two algorithms: BIPACE (bidirectional best hits peak assignment and cluster extension) and CeMAPP-DTW (center-star multiple alignment by pairwise partitioned dynamic time warping).
• Utilizing BIPACE for individual multiple alignment and as a preprocessing step for CeMAPP-DTW.
• Evaluating algorithm performance on GC-MS datasets from Leishmania parasite and Triticum aestivum (wheat).

Main Results:

• BIPACE demonstrates high precision, recall, and low false positive rates in peak assignments.
• CeMAPP-DTW identifies a significant number of true positives, with performance further enhanced when guided by BIPACE.
• Combined approach of BIPACE and CeMAPP-DTW offers superior alignment accuracy for complex GC-MS data.

Conclusions:

• The developed algorithms, BIPACE and CeMAPP-DTW, provide effective solutions for multi-dataset retention time alignment.
• BIPACE enhances the accuracy and reduces false positives, while CeMAPP-DTW improves true positive identification when used together.
• The algorithms are integrated into the OpenSource Maltcms software framework, promoting accessibility and further research.