Skip to content

Processing

The Processing tab is where MINT performs production extraction for your workspace, generating the final results table used by downstream analysis and exports.

Processing

Tip: Click the help icon (small "i" symbol) next to the "Processing" title to take a guided tour of this section.

Processing Workflow

  1. Set Scope and Resources: Click RUN MINT. Choose bookmarked-only mode, recompute/fitting, and compute resources.
  2. Run Extraction: MINT computes chromatograms, then results, then optional EMG fitting.
  3. Export Or Clean Up: Download results or remove selected/all result rows.

1. Set Scope and Resources

RUN MINT opens the execution modal for Processing.

Processing

  • Bookmarked Targets Only: Restrict execution to bookmarked targets.
  • Recompute results: Recalculate already existing result rows.
  • Enable fitting (EMG): Run optional peak fitting after result extraction.
  • Resources: Configure CPU, RAM, and batch size.
Run-time behavior and defaults
  • CPU/RAM defaults are auto-detected from your machine.
  • Batch size is auto-calculated from resource settings.
  • Processing runs in staged phases with progress updates (Chromatograms -> Results -> optional Peak Fitting).
What 'Recompute results' affects
  • Recompute targets the results stage (and fitting stage, when enabled).
  • During Processing runs, chromatogram generation uses the Processing pathway and is not the explicit MS1/MS2 recompute toggle used in Optimization.

2. Results Table

After completion, results becomes the primary working dataset for review and downstream steps.

Processing

  • Target filter dropdown: Quickly narrow visible targets.
  • Server-side filtering/sorting: Efficient browsing for larger result sets.
  • Optional fitted metrics: Available when EMG fitting is enabled.
Core result columns
  • Core extraction metrics such as peak_area, peak_max, peak_mean, peak_rt_of_max.
  • RT-alignment fields such as rt_aligned, rt_shift.
  • Optional EMG fit fields such as peak_area_fitted, peak_rt_fitted, fit_r_squared, fit_success.
Processing filter semantics
  • Results shown in this page are constrained to samples marked use_for_processing = TRUE.
  • The target dropdown controls which compounds are rendered in the table.

3. Exporting Data

Use DOWNLOAD RESULTS to export either full (tidy/long) outputs or dense matrices.

  • All Results (tidy/long): Column-selectable export of raw results rows.
  • Dense Matrix (wide/pivot): Pivot by selected row/column/value fields.
All Results export behavior
  • MINT validates selected columns against supported export fields.
  • Export prefers a workspace backup file when available for faster delivery.
  • For large files, MINT switches to direct Flask file download to avoid browser/base64 bottlenecks.
Dense matrix validation
  • Requires valid row/column/value selections.
  • Prevents invalid combinations (for example identical row and column keys).

Managing Results

Use the Options menu to clean up result subsets or reset the table.

  • Delete selected results: Removes selected (peak_label, ms_file_label) pairs.
  • Clear table: Removes all rows from results.
Deletion semantics
  • Delete selected removes only selected target/sample pairs.
  • Clear table removes the complete results table content.
  • Deletions invalidate the results backup snapshot so exports reflect current DB state.

Results Backup Snapshot

After a successful Processing run, MINT writes a workspace-local CSV backup snapshot:

  • Path: <workspace>/results/results_backup.csv
  • Purpose: Faster exports and resilience/recovery.
  • Update behavior: Regenerated after runs; invalidated when results are deleted so the next export rebuilds cleanly.

SCALiR (Optional)

SCALiR is an optional absolute-quantification workflow layered on top of Processing results.

Processing

  • Provide a standards file.
  • Fit calibration curves.
  • Export concentrations and (optionally) calibration plots.