Spectral Code Index (SPECOIND): A General Infrared Spectral Database Search Method

Abstract

A new spectral code that can be used by Relational Database Management Systems (RDBMS) as an index for infrared (IR) spectra searches in Relational Database (RDB ) is presented and its suitability is evaluated. Spectral codes are constructed for all spectra in the database as the spectral indexes and three query strings are created with the same theory used for the creation of the index code for the query spectrum . Some effects of parameters used to create index strings and query strings are discussed. All spectra l searches are accomplished in structured query language (SQL) approach and the utilization examples of SQL have been shown. The sequential application of this procedure can reduce the original library of about 18 000 spectra to a few spectra that can be used as referen ces for subsequent detailed comparison. The software developed for the proposed system is particularly suitable for spectral search and structure interpretation.Index Headings: Infrared spectroscopy; Spectral database; Spectral code index; Index; Code string. INT RODUCTIO NBecause of the limitations of the knowledge of the natural world or the calculation power, empirical research methods, not calculation, play a signi cant role in the chemical research eld. The database is the data infrastructure of any empirical research. The spectra are a very important resource in chemistry applications and research domains. Searching spectral libraries of known compounds is one of the m ost important tools in spectral interpretation and structure elucidation.1-4 The key to using this tool is that the spectral collection m ust be large enough to give a decent probability of a match. In 1991, Warr 5 noted that ''no single spectral database contains m ore than app ro xim ately 1 00 00 0 co m pou nds ev en though there were ten m illion known chemical compounds in the CA Registry System. There were 20 -30 times as many hard-copy libraries of spectra as there were computer-readable spectral libraries''. The situation is changing. For example, Bio-Rad's Informatics Division 6 announced that it had the largest collection of infrared (IR ) data in the world of over 220 000 spectra of pure organic and commercial com pounds. Before the 1990s, most of the spectra were stored in printed format, 7-12 and some of the handbooks have indexes on compound name, molecular form ula, chemical class, numerical, and SpecFinder Code. Nowadays, there are m ore and more digital and on-line spectral databases that can provide more comprehensive search tools and higher quality spectra. NIST Chemistry WebBook, 16 and Integrated Spectral Data Base System for Organic Compounds (SDBS) 17 are fam ous on-line free-access public databases. Because spectral databases are very useful, not only big companies but also many research groups have given attention to creating and m anaging databases. As the power of personal computers has grown stronger, some publications can be found that discuss the application of personal computers to creating an IR spectral database and a spectral ret...

Dual-Principal Component Analysis of the Raman Spectrum Matrix to Automatically Identify and Visualize Microplastics and Nanoplastics

“…In Figure b, when the three PCA spectra are compared with eight Raman spectra, the situation is much more complicated, and an algorithm approach is thus preferred. An algorithm-based comparison is to recognize a digital value to indicate the similarity or difference, among a suspected spectrum and the standard spectra, akin to an index . Fortunately, the correlation of PCA analysis can be used for this comparison and justification.…”

“…Taking the PCA spectrum of PC1 from the first round of PCA for example, we compare it with the eight standard spectra, in order to identify the type of plastic . To avoid the comparison bias and to increase the accuracy, we pretreat the spectrum, as discussed above.…”

“…Then, we apply a dual-PCA process that involves (i) an initial PCA analysis (first round) to create the PCA spectra and generate the PCA intensity images and (ii) a second round of PCA, by combining the PCA spectra created in the first round of PCA with eight standard spectra of the common plastics, to automatically and digitally assign the principal components (of the first round of PCA) to the suspected items via the correlation matrix (of the second round of PCA), akin to an index. 21 To validate the effectiveness of this improved method, we test it first on a mixture of two virgin microplastics and then on a sample collected during lawn trimming in our gardens. The results from this study might be useful in facilitating the development toward machine learning for the analysis of microplastics and little-known nanoplastics.…”

Raman imaging combined with an improved PCA/algebra-based algorithm to capture microplastics and nanoplastics

“…After the 1 st round of PCA, the PCA spectrum, such as of PC1, is combined with 10 standard spectra, in order to identify it, akin to indexing. 31 To avoid the comparison bias and to increase the accuracy, we pre-treat all the PCA spectra and Raman spectra, using techniques including cosmic ray removal, smoothening, baseline correction, interpolating to adjust the wavenumber ( x -axis) and normalising the intensity ( y -axis). Most of these pre-treatments can be conducted using either the operational software (WITec Project, such as cosmic ray removal) or Origin.…”

Algorithms Research of near Infrared Spectral Database of Apples

“…In general, there are mainly two kinds of SMA. The first kind matches spectra by the similarity of spectral peak information (SMA-P), [7][8][9] so peak detection is an essential step for subsequent analysis. The thresholds of signal-to-noise ratio, peak height, peak width and peak shape index are commonly used for peak detection.…”