Computer-assisted structure elucidation of organic compounds III: Automatic fragment generation from13C-NMR spectra

Computer-Based Structure Determination: Then and Now

“…Several structure elucidation systems, e.g., ACCESS, 73 CSEARCH, 74 EPIOS, 75 and SpecSolv, 76 have tightly linked spectrum interpreters and structure generators. Each utilizes a 13 C NMR interpretive library search to infer the presence of explicitly defined substructures contained in a library of such fragments created from a database of assigned 13 C NMR spectra.…”

Predicting phosphorus NMR shifts using neural networks

“…Tables IV and V show the number of compound vectors we have allocated to each of the three sets. In each table, column 3 shows the number of substructure code types in the 3c3v class. The range that the shifts of the 3c3v class spans in parts per million is given in column 4.…”

General Characteristics of Good-List and Bad-List Entries for Structure Generators from Spectra

“…This does not necessarily lead to a loss of valid structures, if the database is used for substructure identification (module 1). [27][28][29] However, 100% reliability is not granted if the whole generation process relies on a database, although such systems can be impressively efficient in favorable cases 30 if signal assignments are updated during the structure generation. 15,30 In this work, the reliability and efficiency of fragments as good-list or bad-list items are investigated with a view to designing a rule-based system.…”

“…Systems relying on databases of measured spectra are not comprehensive for obvious reasons. This does not necessarily lead to a loss of valid structures, if the database is used for substructure identification (module 1). − However, 100% reliability is not granted if the whole generation process relies on a database, although such systems can be impressively efficient in favorable cases if signal assignments are updated during the structure generation. , …”