Raw Data Files for EI Low Resolution in the Identifications Folder

Importance of the topic

The dataset and file products generated by NIST (low-resolution EI GC/MS) during chromatogram processing provide an essential audit trail for deconvolution and library-search identifications. Understanding these intermediate files is important for improving result transparency, ensuring reproducibility, diagnosing deconvolution behavior in complex mixtures, and enabling downstream comparative and multivariate analyses (difference-finding, PCA, supervised classification) across samples in QA/QC, environmental, food, forensic and research workflows.

Objectives and study overview

This material documents the role and contents of the NIST/AMDIS-style output files stored in the Identifications folder (example: Grob analysis in EI, no retention index). The primary goals are to: (1) explain what each output file records, (2) show how deconvoluted spectra are produced and searched, and (3) evaluate whether these files can be used for sample-to-sample comparisons or integrated into external, user-written applications for difference analysis and multivariate statistics.

Methodology and instrumentation

The workflow described applies to low-resolution electron ionization (EI) GC–MS processed with NIST deconvolution/library-search software. The processing chain is: raw GC/MS data → extracted ion traces → component detection/deconvolution → deconvoluted spectra → library searching. The software generates an experimental results folder that contains intermediate and final products documenting each processing step and parameters used.

Instrumentation used

• Low-resolution GC–MS with electron ionization (EI).
• NIST processing tools (NIST/AMDIS-style deconvolution and NIST library searching; referenced as NIST26 in the source material).

Main results and discussion

• Key output file types and roles:
  • .TIC — total ion chromatogram: broad overview of chromatographic signal but limited for compound-level comparisons.
  • .ELU — elution/profile files: describe ion elution behavior and peak shapes useful to inspect chromatographic separation and coelution.
  • .FIN (example GROB.FIN) — final deconvoluted spectra: one cleaned spectrum per resolved component produced by deconvolution; the primary input for library searching.
  • .TSV (example GROB.tsv) — tab-delimited summary: retention times, match factors (library scores), identified compounds, intensity/area values; most practical for sample comparisons and reporting.
  • .run — processing configuration: deconvolution settings, detection thresholds, library search options; critical for reproducibility and interpreting results.
• Practical insight from the dataset: deconvolution dramatically reduces the number of spectra searched. For the example Grob run, only 91 deconvoluted component spectra were searched rather than every scan. This highlights that NIST/AMDIS summarizes thousands of scans into a small number of chemically meaningful spectra prior to library querying.
• Comparative analysis potential: these output files can support sample-to-sample comparisons. The recommended strategy is to match components by a combination of retention time, library identification, and major ions or spectrum. Useful comparisons include presence/absence of compounds, peak area/intensity changes, relative abundance shifts, appearance/disappearance of deconvoluted components, and changes in library match scores.
• Best files for comparison:
  • .TSV — best starting point for tabulated component-level comparisons and easy import into statistical tools.
  • .FIN — use to compare deconvoluted spectra directly (spectral similarity metrics, cosine scores).
  • .ELU — use for checking peak shape and coelution artifacts that may affect quantitation or deconvolution reliability.
• Limitations and cautions:
  • Deconvolution is parameter-dependent: differences in .run settings, noise thresholds, or algorithm versions can change the set of deconvoluted components and match scores.
  • Retention time alignment across runs is required before automated matching; small RT shifts can break direct RT matching and must be corrected (alignment, warping).
  • Quantitative comparisons require consistent sample preparation, injection/response normalization, and ideally internal standards; raw peak intensities from deconvolution may not be directly comparable across runs without normalization.

Benefits and practical applications of the method

• Transparency and auditability: intermediate files document how spectra were constructed and why library identifications were proposed.
• Reproducibility: .run files store processing parameters enabling repeatable deconvolution and searches.
• Flexible downstream analysis: tabular and spectral outputs can be imported into custom pipelines for differential analysis, clustering, PCA, or machine-learning classification (good vs bad sample screening, trend detection).
• Data reduction: deconvolution reduces large chromatographic datasets to a concise set of chemically-relevant spectra suitable for targeted reporting or multivariate comparison.

Future trends and potential uses

• Closer integration of deconvolution with library searching and searchable audit metadata (e.g., NIST26 with integrated deconvolution) to streamline workflows and reduce manual inspection.
• Adoption of spectral alignment and library-agnostic approaches (e.g., spectral networking, mass spectral embeddings) combined with classical multivariate methods (PCA, PLS-DA) for robust sample classification and anomaly detection.
• Machine-learning models trained on deconvoluted spectra and metadata to automate component matching and flagging of inconsistent deconvolution results across batches.
• Standardization of output file formats and richer metadata to facilitate exchange and use in external, user-written applications for comparative statistics and visualization.
• Extension of these approaches to high-resolution EI and LC–MS/MS contexts with integrated deconvolution and accurate-mass constraints.

Conclusion

The NIST-generated intermediate files produced during low-resolution EI GC–MS processing are a valuable resource for understanding deconvolution behavior, verifying library identifications, and enabling sample comparisons. The .TSV and .FIN files are especially useful starting points for automated difference-finding and multivariate analyses, while .run and .ELU files help ensure reproducibility and diagnose chromatographic/deconvolution artifacts. Proper normalization, retention-time alignment, and consistent processing parameters are essential for reliable cross-sample comparisons. With continued improvements in integrated deconvolution/library tools and standardized outputs, these files will increasingly support advanced statistical workflows and machine-assisted interpretation.

References

• James Little. Mass Spec Interpretation Services. Raw dataset and notes from course material, April 24, 2026.
• NIST (referred to as NIST26) deconvolution and library-search workflow (as described in source material).