Spectra Prediction Methods

Spectral prediction is required especially in the case of characterization and structure elucidation of large complex molecnles, such as natural products [30], A complete one-to-one correspondence for the assignment of the peaks in the spectra is not possible from the experimental spectra. Prediction is also required in the case of mechanistic nnderstanding for synthetic organic chemistiy. Many methods have been developed to predict spectrum, given stmctural information. 

Empirical methods employ additive rules usually called as incremental methods [31]. 

Semiempirical methods are based on the classical concepts of inductive and resonance contributions and employ molecular mechanics force fields [32]. 

It is the most widely employed approach in most software, for instance, Advanced Chemistry Development, Inc. (ACD/Labs). It is faster because three-dimensional (3D) geometries are not determined only matching with stored chemical shifts is 

Machine learning methods such as artificial neural networks are employed for both small molecules and protein stmcture prediction [35]. 

---

Write a brief essay on known spectra prediction methods. Highlight the advantages and disadvantages of each method. 

Analysis of the CD spectrum has yielded values of 14% a helix and 31 % p strand, with a possible increase in helix content observed with increase of temperature (Loucheaux-Lefebvre et al., 1978). In a more recent study (Ono et al., 1987), a lower fraction of a helix was calculated, but the results vary with the method of calculation. Structure prediction methods have also been applied to this protein and have given results that encourage the view that K-casein has a number of stable conformational features. Loucheaux-Lefebvre et al. (1978) applied the Chou and Fasman (1974) method and predicted an a-helical content of 23%, with 31% P strand and 10% p turns. Raap et al. (1983) preferred the method of Lim (1974) to predict a-helix and P-strand content, because the method of Chou and Fasman, as published in 1974, was considered to overpredict these elements (Lenstra, 1977). They also tested their predictions for the structure about the chymosin-sensitive bond using the later boundary analysis method... 

As stated above, spectrum prediction is based on the correlation between structural environments and their corresponding chemical shift values. In order to simulate spectra by the methods described adequate reference material is necessary, which is not given in some journals, frequently, instead of assignments only peak-... 

Spectrum prediction is a frequently used technique during the structure elucidation process, but a detailed inspection of the results is necessary. Some programs offer the possibility to use different algorithms for spectrum prediction (usually HOSE code technology and neural networks). In such a situation both methods should be applied and the results obtained should be carefully compared [22]. At least in the case of different predictions a further critical evaluation of the result should be an obligation. 

Experimental < s measured in reflected shock waves with the spatially-integrated CO-0 flame spectrum emission method over a range of temperatures and for different values of rj, and normalized by 2[02], are displayed in Fig. 2.8. In accordance with equation (2.13), the data separate only with rj at higher temperatures. Variation of /2[02] with total gas concentration, predicted by equations (2.11) and (2.12) is seen inr] = 0-33 mixture data at the lowest temperatures. Also shown are recent infrared measurements which have extended the previously limited = 10 results up to nearly 2200 K, while yielding excellent agreement with the CO-0 data. 

Nuclear magnetic resonance (NMR) spectroscopy is the most informative analytical technique and is widely applied in combinatorial chemistry. However, an automated interpretation of the NMR spectral results is difficult (3,4). Usually the interpretation can be supported by use of spectrum calculation (5-18) and structure generator programs (8,12,18-21). Automated structure validation methods rely on NMR signal comparison using substructure/ subspectra correlated databases or shift prediction methods (8,15,22,23). We have recently introduced a novel NMR method called AutoDROP (Automated Definition and Recognition of Patterns) to rapidly analyze compounds libraries (24-29). The method is based on experimental data obtained from the measured ID or 2D iH,i C correlated (HSQC) spectra. 

The experimental data, also in gas phase, is from Ref. [41]. In the region below 7 eV, furan shows a series of Rydberg states over-imposed to a broad band. The nuclear-ensemble method provides a good qualitative prediction of the spectrum. The intensity and the shape of the broad band are in very good agreement with the experiment. The energy shift is caused by the electronic structure method (see Sect. 5.1), rather than by the spectrum simulation method itself. 

The HOSE code (hierarchically ordered spherical description of environment) method was introduced by Bremser in 1978 and is widely used in commercially available spectrum prediction programs. The fundamental idea is to convert the connectivity table into a linear notation and to correlate this substructure description with the corresponding chemical shift value. 

The three basic methods and their extensions described so far are mainly used for spectrum prediction in this case the correlation of a particular carbon atom within a certain environment with its specific chemical shift value is necessary and must be accessed during spectrum prediction. During the spectrum prediction process the query structure is analyzed in terms of single carbon atoms with their environments this information is checked against a lookup table and for each carbon a certain shift value is predicted. The final result is obtained by repeating this process for each carbon atom independently. 

The result of spectrum prediction is also quite dependent on the reference data collection used as the knowledge base. From our own experience with the CSEARCH database system usually deviations of 1-3 ppm can be expected, depending on the method used for calculation and the representation of the query structure within the database. 

Spectrum prediction is based on a 1 1 correlation between structural properties and spectral features therefore well-assigned NMR spectra are necessary. The chemical literature offers a large variety of spectral information having different levels of quality. In some journals most of the C NMR spectra have been assigned by two-dimensional NMR methods, whereas other Journals use unassigned peak lists as given by the NMR equipment for structure verification - this information is completely useless for spectrum prediction using the methods described here. 

In structure elucidation using NMR techniques, spectrum prediction or calculation software has been playing a significant role for many years. Table 6 summarizes common methods used in practical work. 

Many attempts have been made in the recent years to calculate IR spectra on the basis of semiempirical or ab initio methods. Since the computational times required are rather high, the practical use of these techniques is limited. In both NMR and IR spectroscopy, spectrum prediction techniques based on structure-oriented libraries are the methods of choice. 

The verification module uses spectrum prediction tools to reduce the solution space. A reduction of up to 99% is achievable, if all spectroscopic methods are used, because each spectroscopic method has specific properties to describe structural features. 

There are three types of capability of major importance in structure determination. Spectrum interpretation is the process by which spectral data are reduced to structural information, which is usually expressed in terms of substructures predicted to be present or absent in the compound under study. Structure generation serves to exhaustively generate all molecular structures compatible with this structural information. Spectrum prediction and comparison are important in evaluating the relative merits of the structures in a set of plausible alternatives. (Suitable methods for the input, representation, storage, manipulation, comparison, and graphic display of chemical structures and spectra may also be necessary components of the computer implementation of these three capabilities, but a description of them is beyond the scope of this article.)... 

Applications of spectrum prediction to the evaluation of candidate structures have some special requirements. First, comparisons are to be between predicted and experimental spectra, not relative comparisons between predicted spectra therefore, the predicted spectrum of a compound must closely approximate its experimentally determined spectrum. Second, the methods must be applicable to larger, complex, highly functionalized compounds as well as smaller, simpler ones. Third, spectrum prediction must be sufficiently refined to yield spectral distinctions between isomeric compounds that possess structural similarities (structural building units and constraints), which at times can be substantial. If they are to be of value, the techniques should be more discriminating than those used in spectrum interpretation. Finally, since at times there may be many structures whose spectral properties are to be predicted, the methods should be computationally efficient. 

A number of factors in addition to program versatility and reliability influence the quality of spectrum prediction. Currently, the information input to most spectrum prediction programs relates to connectivity, not to three-dimensional structure. Spectroscopic methods differ in the extent to which spectral properties are influenced by the stereochemistry of the compound, e.g., NMR spectra are quite sensitive, mass spectra are not. Furthermore, depending on the spectroscopic method, variations in acquiring the data (e.g., the particular spectrometer and solvent used) can also lead to spectral differences. For these reasons, spectrum prediction programs are generally used to rank, rather than eliminate candidate structures. 

The most common approaches to predicting spectra are based on empirical modeling, linear additivity, database retrieval, rule sets, and semiempirical methods. The availability of large spectral libraries has proved to be a valuable resource in these studies. Although ab initio theory relating to spectrum prediction is well advanced, the theoretical equations necessary for application to real-world structures are large and complex... 

---