13
New Investigator Tools for Finding Unique and Common Components in Multiple Samples with Comprehensive Two-Dimensional Chromatography
Qingping Tao;1 1
2 3 Stephen E. Reichenbach;1,2 Chase Heble;1
GC Image, LLC, Lincoln NE, USA University of Nebraska, Lincoln NE, USA Zoex Corporation, Houston TX, USA
Zhanpin Wu;3
Comprehensive two-dimensional chromatography is a powerful technique for highly effective chemical separations of complex mixtures and increasingly is used for cross-sample analyses such as sample classification and biomarker discovery. These techniques, such as GCxGC and LCxLC, produce large data sets that are rich with information, but highly complex, and that require automated processing with robust methods. An important challenge is to select a few markers that can be used effectively for clustering and classifying multiple samples. A newly developed workflow and associated tools allow analysts to detect common and unique compounds across many samples with specialised detection and identification constraints that use chromatographic and mass spectral information to distinguish marker compounds. In addition, new visualisation tools for multi-classification methods provide not only metric values, but also instructive predictions as to which features are effective for distinguishing samples.
Introduction
Untargeted cross-sample analyses such as sample classification and biomarker discovery require separating, quantifying, and identifying a large number of compounds in chemically rich samples and then relating the complex compositions across samples and sample classes. Advanced chromatography, mass spectrometry, and statistical data analysis methods can be combined to address this challenge [1]. In particular, separations performed with comprehensive two-dimensional chromatography (such as GCxGC and LCxLC) provide much greater separation capacity and signal-to-noise ratio than traditional one-dimensional chromatography [2,3]. Coupled with high- resolution accurate mass spectrometry, comprehensive two-dimensional chromatography is a powerful analytical solution. However, the large and complex data sets produced also present challenges for data analysis.
The InvestigatorTM framework (GC
Image, Lincoln NE, USA) developed previously analyses data from multiple samples to extract a feature template that comprehensively captures the pattern of
Figure 1. Automated Investigator Workflow. The workflow detects and aligns all compounds from multiple chromatograms, and extracts feature vectors for untarged cross-sample analyses.
peaks detected in the retention-times plane [4, 5]. Automated feature template extraction, as outlined in Figure 1, is performed by: (1) matching peaks to construct a pattern of alignment peaks that can be reliably matched across chromatograms; (2) aligning and combining chromatograms across samples to create a composite chromatogram; and (3) detecting peak-regions observed in the composite chromatogram. Then, for each sample chromatogram, the extracted feature template is transformed to align
with the detected peak pattern and used to generate a set of feature measurements from transformed peak-regions for cross-sample analyses. The approach avoids the typically intractable problem of comprehensive peak matching and can produce feature templates with thousands of features.
The result of the Investigator framework is a feature database with three data dimensions: the chemical features extracted (i.e., peaks and peak-regions); the various attributes measured for each feature, such as retention
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68
