« Back
Automated Deformulation of LC/MS and GC/MS Data through Database Searching
Poster Title: Automated Deformulation of LC/MS and GC/MS Data through Database Searching
Submitted on 11 Jul 2018
Author(s): Anne Marie Smith, Richard Lee, Artsiom Piatrouski, Andrey Paramonov, Vitaly Lashin
Affiliations: Advanced Chemistry Development, Inc. (ACD/Labs); 8 King Street East, Suite 107; Toronto, ON, Canada; M5C 1B5
This poster was presented at 66th American Society for Mass Spectrometry Conference
Poster Views: 1,641
View poster »

Poster Information
Abstract: Automated Deformulation of LCMS and GCMS data through Database Searching

As the capabilities of mass spectrometers (MS) progress with improved resolution and mass accuracy, the demand for analysis of samples that are more complex have emerged. Some of the sample complexity can be reduced with the use of chromatography; however, co-elution of components is nearly unavoidable.

While the use of MS instruments with greater capabilities can be used to more accurately determine elemental composition of components, this still lacks structural information. Here we describe an extension of a new chromatographic deconvolution algorithm based on ion threads to show components throughout the sample analysis while automatically searching databases to aid in structure identification.

The feature-finding algorithm (IX 2.0) was used to generate component ion threads and mass spectra that could automatically be searched against a database. Known samples were analyzed for the accuracy at which the software was able to identify structures, including samples from both liquid chromatography (LC) and gas chromatography (GC) analysis. LC-MS samples were searched against a database containing ESI MS2 spectra, while GC-MS samples were searched against a database containing EI MS spectra. Databases searched included a composite of commercially available databases in a *.nd9 format and internal user-created databases (*.cfd) to improve identification of commonly found components within a user’s common chemical space.

Preliminary Data
The component detection and automated searching algorithm was tested for its accuracy in identifying components found in a variety of samples and databases. The newly developed feature uses the extraction of ion threads for identification of components of a sample. The processing routine involved extraction of ion threads, peak integration, grouping of spectral features to generate a component mass spectrum, search component spectra and updating the project with the component and search results. In the case of GC-MS data, the software automatically searched a database(s) for matching EI MS spectra. Due to the lack of reproducibility of MS spectra in LC-MS analysis, the MS2 spectra of found component spectra were automatically searched in user-curated database(s).

The new database search feature populated the table of components with the addition of the structure column populated with the top database hit. The software annotated spectra and populated the table with possible fragment, and adduct ions. The data was further interrogated by allowing an expert user to manually review the hit results, allowing the end user to select from any of the returned hits.
The presented new feature provided an important advantage in identifying structures—presenting an extensive, unbiased and relevant list of structures to researchers.

Novel Aspect
Automated deformulation of complex samples through spectral searching for structure identification of features found in LC-MS and GC-MS datasets.
Summary: Describing our new IntelliXtraction Component Recognition (IXCR) 2.0 deconvolution algorithm, which provides an important advantage in identifying structures in unknown LC/MS and GC/MS datasets via spectral searching — presenting an extensive, unbiased, and relevant list of structures to researchers, easing resource strain for deformulation of complex MS samples.Report abuse »
Ask the author a question about this poster.
Ask a Question »

Creative Commons

Related Posters

Cutting-edge Bioinformatics Solutions for Proteomics, Metabolomics, & Proteins
Creative Proteomics

Targeted Dereplication of Microbial Natural Products by High-Resolution MS and Predicted LC Retention Time
Chervin J,1 Stierhof M,1 Tong MH,1 Peace D,1 Hansen KØ,2 Urgast DS,1 Andersen JH,2 Yu Y,3 Ebel R,1 Kyeremeh K,4 Paget V,5 Cimpan G,5 Van Wyk A,5 McKee M,5 Deng H,1 Jaspars M,1 Tabudravu JN1

New dataflow model for rapid processing of large volumes of LC/UV/MS data
Richard Lee1 Andrey Paramonov 1 , Eugene Vodopianov 1 , Vladislav Solomatin 1

Consistent Detection of Multiplets in 13C Spectra Using Structure Aware Algorithms
Dimitris Argyropoulos, Sergey Golotvin, Rostislav Pol and Vladimir Mikhailenko

Efficient Approaches for Addressing Spectral Ambiguities in Computer Assisted Structure Elucidation (CASE) Systems
Dimitris Argyropoulos, Rostislav Pol, Mikhail Elyashberg and Sergey Golotvin