Analyze a single compound or mixture and review molecule descriptors before prediction.
Prerequisite: Load The Molecule From ChemrytIQ
Before opening ChemrytHPLC, search the molecule in ChemrytIQ by SMILES, InChI, molecule name, or CAS number. Confirm the correct molecule on the ChemrytIQ page, then open the required Chemryt app from that same molecule context so the selected structure is loaded into the app automatically.
What ChemrytHPLC Does
ChemrytHPLC supports molecule drawing, SMILES or molfile input, method metadata, method optimization, gradient controls, backpressure guardrails, predicted chromatograms, chromatographic metrics, and future-ready integration objects.
Capture method template, column chemistry, brand, mobile phases, additive, and chromatographic mode.
Tune objectives, gradient controls, instrument guardrails, method notes, and chromatographic metrics.
Quick Workflow
- Load the molecule from the editor, SMILES field, molfile import, example list, or ChemrytIQ transfer.
- Select the prediction target mode and fill in method metadata such as column chemistry, mobile phases, additive, pH, and mode.
- Use method optimization controls when you need to balance runtime, resolution, robustness, and pressure.
- Preview the gradient and estimated backpressure profile before running the final prediction.
- Run the prediction and review retention, method interpretation, chromatogram, and metrics dashboard.
- Copy or export the integration object when another workflow needs the method-specific payload.
Main Areas
| Area | What to enter or review | When to use it |
|---|---|---|
| Input | Molecule editor, SMILES, molfile, examples, mixture components, and descriptors. | Use to prepare the chemical target. |
| Method | Template, column, mobile phases, additive, mode, pH, gradient, pressure, and instrument guardrails. | Use to model the HPLC method realistically. |
| Results | Predicted retention, chromatogram, metrics dashboard, notes, warnings, and integration object. | Use to decide whether the method is ready for lab scouting. |
Tutorial Notes
- Choose reversed-phase, HILIC, ion-exchange, or another mode based on analyte polarity and ionization behavior.
- Check buffer capacity and sample-solvent mismatch warnings before trusting a retention trend.
- Use a gradient preview for complex mixtures and a simpler isocratic setup for quick single-compound scouting.
- Treat predicted pressure as a guardrail, then verify on the actual instrument and column hardware.
ML Model / Computation Used
| Model or method | What it predicts | Implementation details |
|---|---|---|
| ChemrytHPLC retention-time baseline | Method-specific retention time in minutes. | Hybrid scaffold-split models use RDKit descriptors, Morgan fingerprints, and method features. Routes include RepoRT, SMRT, and mixed joblib models; manifest metrics include MAE about 1.29-1.74 min and R2 about 0.73-0.79. |
| Frontend chromatogram simulation | Peak width, intensity, tailing, baseline, secondary peaks, and visualization. | These are simulation outputs around the retention prediction, not separate trained retention models. |
Good Practice
HPLC predictions help plan methods, but final method acceptance requires standards, suitability injections, robustness checks, and validated instrument conditions.
Reference Used
This Tutorial page was prepared from the ChemrytLabs reference module: ChemrytHPLC.