Prerequisite: Load The Molecule From ChemrytIQ

Before opening ChemrytUVIR, 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 ChemrytUVIR Does

ChemrytUVIR accepts drawn structures or typed SMILES, calculates molecule properties and functional-group summaries, predicts UV-Vis, IR, and mass spectra, displays peak assignments, supports PNG/CSV export, and stores prediction history.

Structure input

Draw in the molecule editor, type SMILES, clear, load examples, or import from ChemrytIQ.

Prediction studio

Choose spectral mode and generate UV-Vis, IR, or mass spectrum output.

Results and history

Review molecule information, messages, peak assignments, chart output, PNG/CSV export, save, and history pages.

Quick Workflow

Draw a molecule or paste a SMILES string, then confirm the molecule information panel.
Select the desired prediction mode for UV-Vis, IR, or mass spectra.
Run prediction and watch the message panel for validation or runtime notes.
Review the predicted spectrum chart and peak assignment table.
Export PNG or CSV when you need the spectrum in reports or downstream analysis.
Save the prediction and use the history page to reopen previous runs.

Main Areas

Area	What to enter or review	When to use it
Molecule	Editor, SMILES, formula, molecular weight, properties, and functional group summary.	Use to confirm the structure before prediction.
Spectrum	UV-Vis, IR, and mass prediction modes with chart rendering.	Use to explore different analytical signatures from the same structure.
Output	Peak assignments, PNG export, CSV export, save, and history.	Use to document and compare prediction runs.

Tutorial Notes

Run one spectral mode at a time and verify molecule properties before comparing multiple spectra.
Use peak assignments to understand why a spectrum looks the way it does, not just to collect peak positions.
Export CSV for quantitative downstream work and PNG for visual reports.
Use history for repeat review, but rerun predictions after structure or model updates.

ML Model / Computation Used

Model or method	What it predicts	Implementation details
ChemrytUVIR IR GNN v1	Dense IR transmittance curve from molecular graph input.	Uses a PyTorch Geometric GCN-style model with RDKit graph features and the artifact chemrytuvir_models/ir/gnn_ir_best.pt. The PHP wrapper falls back to the rule-based IR predictor if the GNN or dependencies are unavailable.
Rule-based UV-Vis, IR fallback, and MS predictors	UV-Vis peaks, IR fallback peaks, and mass-spectrum assignments.	The module also keeps rule/fragment predictors for modes or environments where a trained model is not available; UV-Vis GNN checkpoint location is reserved but not active in the inspected worker.

Good Practice

Spectral predictions are interpretive aids. Confirm identity and purity decisions with measured spectra, calibrated instruments, and validated reference methods.

Reference Used

This Tutorial page was prepared from the ChemrytLabs reference module: ChemrytUVIR.