Application of Fourier transform infrared (FT-IR) spectroscopy to discriminate serotypes of Streptococcus suis

Abstract Streptococcus suis is a major swine pathogen and is considered an emerging zoonotic agent. Serotyping plays a crucial role for epidemiological investigations, pathogenicity assessment, and vaccine strategy. The present study evaluated the applicability of Fourier transform infrared (FT-IR) spectroscopy, using the IR Biotyper system, for the serotype discrimination of 94 S. suis isolates belonging to 13 distinct serotypes. Analysis of the 1300–800 cm −1 spectral region demonstrated the highest discriminatory power. Principal component analysis (PCA)/linear discriminant analysis (LDA) showed clear separation among the tested serotypes, except for serotypes 1/2 and 2. The classifier achieved overall accuracy of 90%, which increased to 96.7% when serotypes 2 and 1/2 were considered as a single category. These findings demonstrate that FT-IR spectroscopy represents a promising, rapid, and cost-effective tool for S. suis serotype discrimination. Its performance, ease of use, and compatibility with machine learning approaches support its potential implementation in routine diagnostic and epidemiological workflows. As previously reported for other clinically relevant bacteria, FT-IR spectroscopy may serve as a valuable complementary method to molecular typing, particularly in settings requiring rapid decision-making and high-throughput analysis.