Classification of deep-sea cold seep bacteria by transformer combined with Raman spectroscopy

Abstract

Raman spectroscopy is a rapid analysis method of biological samples without labeling and destruction. At present, the commonly used Raman spectrum classification models include CNN, RNN, etc. The transformer has not been used for Raman spectrum identification. This paper introduces a new method of transformer combined with Raman spectroscopy to identify deep-sea cold seep microorganisms at the single-cell level. We collected the Raman spectra of eight cold seep bacteria, each of which has at least 500 spectra for the training of transformer model. We compare the transformer classification model with other deep learning classification models. The experimental results show that this method can improve the accuracy of microbial classification. Our average isolation level accuracy is more than 97%.

Raman ConvMSANet: A High-Accuracy Neural Network for Raman Spectroscopy Blood and Semen Identification

“…These model evaluation metrics demonstrate that our proposed Raman ConvMSANet performs well in multiclassification tasks of Raman spectra and has high credibility. 25 Bratchenko et al., 13 Al-Shaebi et al, 32 Chen et al, 33 Sang et al, 34 and Liu et al 27 were constructed and trained and tested on two data sets to evaluate their classification performance using 10-fold crossvalidation. Each network was pretuned based on the two data sets to ensure the network was trained with the optimal parameters.…”

“…It computes the correlation between local features to obtain global feature relations. The first Raman spectroscopy classification network based on the transformer network structure has been proposed to identify deep-sea cold seep microorganisms at the single-cell level . However, the transformer network structure also has shortcomings.…”

Patch-Based Convolutional Encoder: A Deep Learning Algorithm for Spectral Classification Balancing the Local and Global Information

“…In particular, a transformer-based algorithm incorporates a self-attention mechanism and global receptive field 19 and allows the algorithm to identify dependencies between each part of the spectra in parallel. 20,21 However, both LSTM and transformer-based algorithms 22 show poor performance for the first task (Table 1). Derived from AlterNet, 23 ConvMSANet 24 fused the local and global information by combining 1D CNN and multihead selfattention (MSA) mechanism.…”

Fast Raman imaging through the combination of context-aware matrix completion and low spectral resolution

“…As a non-destructive and label-free technique, Raman spectroscopy has a wide range of applications, such as disease diagnosis, drug classification, material characterization and imaging of biological samples. 1–10 However, due to its inherently small scattering cross section, spontaneous Raman scattering is weak, typically orders of magnitude weaker than fluorescence, and requires exposure times of several seconds or even minutes in extreme cases. Thus, while it enjoys wide application as a “point measurement” technique, where single spectra are acquired, it has so far rather limited application in biomedical imaging.…”