Blank measurement based time-alignment in LC-MS

TL;DR

This paper introduces a blank measurement based time-alignment (BBTA) method for HPLC-MS data that is faster, simpler, and more accurate than existing warping techniques, requiring no peak detection or internal standards.

Contribution

The paper presents a novel BBTA approach that significantly reduces computational time and complexity for time-alignment in LC-MS data, outperforming traditional warping methods.

Findings

BBTA is over 1000 times faster than correlation optimized warping.

BBTA requires no peak detection or internal standards.

BBTA achieves comparable or better accuracy in time-alignment.

Abstract

Here are presenting the blank based time-alignment (BBTA) as a strong analytical approach for treatment of non-linear shift in time occurring in HPLC-MS data. Need of such tool in recent large dataset produced by analytical chemistry and so-called omics studies is evident. Proposed approach is based on measurement and comparison of blank and analyzed sample evident features. In the first step of BBTA procedure, the number of compounds is reduced by max-to-mean ratio thresholding, which extensively reduce the computational time. Simple thresholding is followed by selection of time markers defined from blank inflex points which are then used for the transformation function, polynomial of second degree, in the example. BBTA approach was compared on real HPLC-MS measurement with Correlation Optimized Warping (COW) method. It was proved to have distinctively shorter computational time as well as lower level of mathematical presumptions. The BBTA is computationally much easier, quicker (more then 1000x) and accurate in comparison with warping. Moreover, markers selection works efficiently without any peak detection. It is sufficient to analyze only baseline contribution in the analyte measurement with sparse knowledge of blank behavior. Finally, BBTA does not required usage of extra internal standards and due to its simplicity it has a potential to be widespread tool in HPLC-MS data treatment.