Prerequisite: Load The Molecule From ChemrytIQ

Before opening ChemrytNMR, 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 ChemrytNMR Does

ChemrytNMR supports single-molecule and multi-molecule prediction, structure editing, 1H and heteronucleus workflows, conformer averaging, 2D experiments, spectrum display options, shift tables, workflow actions, analytics, and exportable reports.

Prediction mode

Choose single molecule or multi-molecule sample simulation before building the workflow.

Structure and sample

Use the molecule editor, SMILES import, ChemrytIQ handoff, or sample builder.

Advanced actions

Run prediction, server checks, rescale, broadening, peak picking, deconvolution, JCAMP import, overlays, qNMR, impurities, and reports.

Quick Workflow

Choose single-molecule or multi-molecule mode based on the sample you want to simulate.
Draw or paste the molecule, then confirm identity, formula, generated SMILES, and sample list when applicable.
Select nucleus, solvent, field strength, display scale, and optional qNMR or advanced prediction settings.
Run NMR prediction and inspect the simulated spectrum, predicted shift table, and atom-to-peak mapping.
Use spectrum options, conformer averaging, 2D experiment panels, interpretation, or molecule-wise breakdown as needed.
Run workflow actions such as peak pick, deconvolution, overlay compare, ASV score, qNMR, impurity review, or export report for deeper analysis.

Main Areas

Area	What to enter or review	When to use it
Sample setup	Mode switch, molecule editor, SMILES import, sample builder, selected sample list, and identity summary.	Use to define what the spectrum represents.
Single Molecule vs Multi Molecule workflow	Single Molecule uses one drawn or pasted structure as the prediction subject. Multi Molecule opens the sample builder so users can add several structures, label them, set component ratios, enable or disable entries, bulk import molecules, duplicate entries, and load a demo mixture.	Use Single Molecule for assignment of one compound, atom-to-peak review, 2D maps, and clean structure verification. Use Multi Molecule for mixture, formulation, impurity, solvent, coformer, or reaction-sample screening where the displayed spectrum should represent more than one component.
Prediction settings	Nucleus, solvent, field, spectrum options, 2D experiments, and advanced options.	Use to match expected NMR acquisition context.
2D Experiment dropdown	Choose the 2D NMR map to request with the prediction. For 1H workflows, available choices include COSY, NOESY, ROESY, HSQC, and HMBC. For 13C workflows, available choices are HSQC and HMBC. Other nuclei disable unsupported 2D experiments.	Use when you want topology-based cross-peak support in addition to the 1D spectrum and shift table. Select the experiment before running prediction so the backend can return the matching 2D map.
Workflow Action dropdown	Choose the post-prediction task to run: Predict NMR, Check Server, Rescale Field, Apply Broadening, Peak Pick, Deconvolution, FID FFT, JCAMP Import, Overlay Compare, ASV Score, qNMR, Impurities, Isomers, Relaxation, Cache Key, Empirical Correct, Batch Predict, or Export Report.	Use after the structure and acquisition settings are ready. The selected action controls what the Run Workflow Action button sends to the NMR backend and what result panel is displayed.
Results	Spectrum, predicted shifts, conformer averaging, interpretation, 2D maps, workflow results, and reports.	Use to review assignments and document the run.

Tutorial Notes

Start with 1H or 13C prediction for the clean single-molecule case before enabling advanced workflow actions.
Use atom hover and shift table mapping together; do not interpret isolated peaks without structure context.
For mixtures, verify component list and relative amounts before comparing combined spectra.
Use Single Molecule mode when the question is: what spectrum and assignments should this one structure produce? Draw, paste, import, or load the molecule, confirm the generated SMILES and identity summary, select nucleus, solvent, field, line width, display scale, and optional 2D settings, then run prediction. Users should expect one simulated spectrum, a shift table, atom-to-peak mapping, optional 2D output, interpretation notes, and workflow actions that operate on that single structure.
Use Multi Molecule mode when the question is: what would a sample containing several structures look like? Switch to Multi Molecule, build a draft molecule in the editor, give it a label, enter a ratio value and ratio mode, then click Add To Sample. Repeat for each component, or use bulk import or the demo mixture to start faster. The sample list can be edited, duplicated, enabled, disabled, or cleared before prediction.
Before running a Multi Molecule prediction, check that at least one molecule is enabled and each enabled component has a ratio greater than zero. ChemrytNMR normalizes enabled component weights and shows the total enabled sample weight. Very large mixtures may render slowly, so keep practical screening samples to about 10 molecules or fewer when possible.
When Multi Molecule prediction runs, ChemrytNMR predicts each enabled component, merges the component peak lists, and displays a Combined Sample Spectrum. Users can show only the combined spectrum, turn on component overlays, keep peak labels visible, and highlight overlap clusters. The result summary reports the merged signal count and the interpretation/breakdown panels explain molecule-wise contributions and possible overlaps.
Switching back to Single Molecule restores the single-structure workspace and cached single prediction state when available. Use this when a mixture peak needs follow-up: inspect the component responsible for the signal in Multi Molecule, then return to Single Molecule for detailed assignment, 2D experiment review, deconvolution, qNMR region setup, or report export.
Use the 2D Experiment dropdown when you want ChemrytNMR to generate a predicted cross-peak map along with the normal 1D prediction. Select the nucleus first, then choose the 2D experiment: 1H enables COSY, NOESY, ROESY, HSQC, and HMBC; 13C enables HSQC and HMBC. If the dropdown hides or disables an option, that experiment is not supported for the selected nucleus.
Choose COSY to inspect short-range proton-proton coupling patterns, NOESY or ROESY to inspect topology/proximity-style proton-proton relationships, HSQC to connect directly attached 1H and 13C environments, and HMBC to inspect longer-range 1H-13C correlations. After selecting the experiment, run the NMR prediction; the 2D panel appears only when topology-supported cross peaks are generated.
In the 2D output, users can expect a map titled with the selected experiment, plotted cross peaks, ppm ranges for both axes, matrix size, relation labels such as 2J/3J HH, 1JCH, or long-range JCH, and any backend warnings. COSY maps can include 1H marginal projections, while HSQC maps include 1H/13C orientation and edited phase-style color hints for CH, CH2, and CH3 classes.
If no 2D panel appears, check that the molecule has the atoms needed for that experiment, that the selected nucleus supports the chosen 2D mode, and that a fresh prediction was run after changing the dropdown. A warning such as no topology-supported 2D cross-peaks means the structure and predicted traces did not produce valid correlations for that experiment.
Use the Workflow Action dropdown when you want a specific follow-up calculation rather than just changing display options. Predict NMR renders the selected nucleus spectrum and shift table; Check Server confirms engine/model availability; Rescale Field and Apply Broadening update the simulated trace using the field MHz or line width inputs.
For spectrum interpretation, Peak Pick reports theoretical signal centers, intensities, multiplet labels, integrations, and atom assignments. Deconvolution resolves overlapped envelopes into predicted component curves and atom-linked transitions. Overlay Compare reports similarity metrics such as cosine score, RMSE, and point count when experimental and simulated arrays are available.
For experimental-data workflows, FID FFT converts a raw FID payload into spectrum points, and JCAMP Import reads JCAMP-DX metadata and spectrum points. qNMR integrates selected regions and reports region count, reference region, scale factor, normalized areas, and molecule-wise regions when mixture predictions are active.
For decision-support workflows, ASV Score estimates agreement between a proposed structure and peak positions, Impurities scans unassigned peaks against solvent/impurity assignments, Isomers ranks candidate structures with diagnostic peaks, Relaxation fits T1/T2 intensity series, Cache Key shows whether the run can be reused from cache, Empirical Correct applies solvent/pH context, Batch Predict submits a job, and Export Report produces a report package with structure, spectrum, peak table, and metadata.
Use JCAMP import and overlay compare when measured spectra are available, then document mismatch notes in the report.

ML Model / Computation Used

Model or method	What it predicts	Implementation details
ChemrytNMR atom GNN shift models	Atom-level chemical shifts for 1H, 13C, 15N, 19F, and 31P workflows.	PyTorch GNN artifacts are stored under chemrytnmr_models. The 1H model reports validation MAE about 0.31 ppm; the 13C model reports validation MAE about 3.06 ppm on nmrshiftdb2-derived labels.
NMR processing and workflow models	Peak picking, overlays, conformer/2D support, qNMR, and report calculations.	These steps combine Python-backed processing, rule/parameter logic, and spectrum post-processing around the GNN shift predictions.

Good Practice

Predicted NMR spectra are assignment support. Confirm structural decisions with measured spectra, reference compounds, solvent standards, integration checks, and orthogonal analytical evidence.

Reference Used

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