Cell misuse and cross-contamination can affect the accuracy of cell research results and result in wasted time, manpower and material resources. Thus, cell line identification is important and necessary. At present, the commonly used cell line identification methods need cell staining and culturing. There is therefore a need to develop a new method for the rapid and automated identification of cell lines.

Raman spectroscopy has become one of the emerging techniques in the field of microbial identification, with the advantages of being rapid and noninvasive and providing molecular information for biological samples, which is beneficial in the identification of cell lines.

In a study published in Computational and Structural Biotechnology Journal, a research group led by Prof. LI Bei and his Doctoral student LIU Kunxiang from Changchun Institute of Optics, Fine Mechanics and Physics (CIOMP) of the Chinese Academy of Sciences (CAS) established a Raman spectral database of gastric cancer cell lines, constructed five different datasets through interception and recombination, and fully analysed the classification results for the different datasets.

Researchers constructed five spectral data sets using the spectral data of Raman spectrum such as full spectrum, fingerprint area, high wavenumber area and Raman background, built a stacking integrated learning model for different data sets - SL-Raman, and realized the identification of gastric cancer cell lines.

The differences in biochemical composition between normal and gastric cancer cells, correctly identified gastric cancer cells, and correctly differentiated gastric cancer cells with different degrees of differentiation were distinguished.

The high accuracy obtained by SL-Raman in identifying the Raman spectra of gastric cancer cells reflects the differences in chemical composition among different cells. This differentiation was the main objective of this study, and a powerful cell line identification technique for classifying Raman spectroscopy data was established.

Overall, SL-Raman is more accurate than most machine learning algorithms and is able to fully utilize all the characteristic information in Raman spectral data.

Raman spectroscopy is a technology that, in contrast to other culture-free approaches (such as fluorescence labeling, magnetic labeling, single-cell sequencing, etc.), can identify bacteria without the need for specialized label creation and is straightforward to adapt to other samples. Raman spectroscopy can be used to identify the differences of biological components in biological samples.

The combination of Raman spectroscopy and SL-Raman can identify normal gastric epithelial cells and gastric cancer cells. SL-Raman has successfully combined the advantages of various machine learning technologies with the data of multidimensional spectral data sets.

This technology provides a new method for analyzing Raman spectral data, because it can achieve high recognition accuracy with a small amount of data.