2D-JCOG: Transforming 1D 1H NMR Spectra into J‑δ Correlation Maps via the Shared Splitting Theorem and Graph Neural Networks
Carlos Cobas · 2026
Read 2D-JCOG: Transforming 1D 1H NMR Spectra into J‑δ Correlation Maps via the Shared Splitting Theorem and Graph Neural Networks by Carlos Cobas (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13261620.txt.
Description
Carlos Cobas presented in 2026 a deep learning method that turns a one-dimensional proton NMR spectrum into a two-dimensional map of couplings. Coupling constants reveal how atoms connect inside a molecule, yet crowded spectra defeat the usual rule-based reading. His framework, 2D-JCOG, rests on a physical rule: multiplets joined by a coupling must share at least one common splitting.
A graph neural network treats each splitting candidate as a node and links partners that the rule allows. Attention layers weigh those links with quantum consistency features, while other layers gather context from whole multiplet regions. Trained on simulated spectra with exact answers, the model recovered most true couplings with strong precision. On real database spectra it again found most couplings, with tiny average error in the values. Cobas positions the tool for routine analysis of moderately tangled spectra.
Catalog record
| CREATOR | Carlos Cobas |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 59 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |