ACD/Structure Elucidator

A separate analysis of oxocyclostylidol has been performed recently by Elyashberg (private communication). Elyashberg, in collaboration with the authors of this chapter and other collaborators, previously reported on the applications of a CASE system, ACD/Structure Elucidator to complex molecules and specifically to natural products [45,46,48-58]. The details of the system and its associated algorithms will not be reviewed here and readers are referred to the previous articles for information. 

Elyashberg applied ACD/Structure Elucidator to the problem using the HMBC, COSY, HMBC, and 1,1-ADEQUATE data as inputs. A number of... 

Application of ACD/Structure Elucidator to Known Alkaloid Structures... 

ACD/Structure Elucidator has matured over the past 6 years to become the most capable commercial CASE system available today. The expert system has been described in detail in previous works." " The first generation system was developed for the elucidation of molecular structures of intermediate-sized organic molecules with 20-25 skeletal atoms using simple ID C NMR spectra. The second pubhcation" " reported on developments allowing the elucidation of structures of large molecules, typical of those found in the world of natural products, and utilizing mainly 2D NMR data as the source datasets. The application and success of the system as a powerful and versatile tool for the structure elucidation of complex chemical structures has been demonstrated in numerous works and will not be discussed in more detail in this review. " " ... 

In another, recent example, the ACD/Structure Elucidator resolved 2-D NMR data on a C31 alkaloid that had several ambiguities in connectivity associated with spectral overlap. Twelve candidates were revealed, of which eleven were ultimately ruled out for violating a variety of constraints. The new compound, named quindolinocrypto-tackieine, was thus solved. ... 

In addition to software tools to help postacquisition processing, software tools to help mass spectral interpretation, particularly MS/MS, have taken new strides as well (Heinonen et al., 2008). One example of such a software tool is the MathSpec program. The details of the MathSpec approach have been explained (Sweeney, 2003). MathSpec software is used in conjunction with MS/MS spectra obtained under high-resolution conditions. The software systematically attempts to assemble possible parts (from the MS/MS fragment data) of the molecule into a rational molecule. Other examples of structure elucidation software include HighChem s Mass Frontier and ACD/Labs ACD/MS Manager (Bayliss et al., 2007). Other metabolite prediction software tools such as Meteor are also being incorporated into LC-MS software as tools to help accelerate metabolite detection and characterization (Testa et al., 2005 Ives et al., 2007). 

ACD/Labs Structure Elucidator v7.0. In 42 min of computer time, the program output 296 structures that were sorted on the basis of the match between predicted shifts for each structure and the observed shift data. On this basis, the structure of quindoline, 6, was first in the output list with a deviation d=336 ppm. The next structure in the output list had a deviation of 5.26 ppm. Applications of the Structure Elucidator program to hundreds of examples have shown that only in rare cases does the actual structure have d>5 ppm. 

When StrucEluc was first introduced, it only allowed 1-D-NMR data with other information for structural elucidation of molecules containing fewer than 25 atoms. A recent enhancement of StrucEluc now involves a variety of 2-D NMR data as well as data from IR and mass spectra.Such improvements have allowed this code to elucidate product molecules containing more than 60 atoms. This program now forms a suite of programs offered under the name ACD/StructureElucidator. 

Since the data obtained by low resolution instruments, like QqQ, are enough simple to interpret if compared with the huge amount of data obtained by a HR instrument, interpretation of a product ion spectrum generated by the QqQ is often not straightforward, because of the complex dissociation mechanisms for formation of fragment ions. However, a structure elucidation software, such as ACD/MS Fragmenter or Mass Frontier, developed to predict the fragment ions formed by Cff), has been used to assist the spectral interpretation of unknown structures. 

In spite of such interesting examples, it still remains an open question how molecular similarity is defined with respect to immunological response, since it is also known that cross-reactivity exists in some instances between STLs and mono- as well as diterpenoids [74], It appears therefore a promising topic for future studies to investigate this issue in more detail (i.e. with large sets of compounds and applying quantitative structure-activity relationship methods) in order to elucidate the general rules by which specificity in ACD cross-sensitivity is determined. 

---