RESUMO
We assessed the ability of Fourier transform infrared spectroscopy (FT-IR) to differentiate three important and morphologically similar Aspergillus species: A. ochraceus and A. westerdijkiae, and A. niger. Fungi were processed by two methods, powdered mycelia and conidiospore-saline solution, and then recorded in a spectrometer. Second derivatives with nine points of smoothing were applied as spectra data pretreatment. Partial least squares regression was used for the species comparison models and a prediction test was used to evaluate the models. The powdered-mycelia methodology correctly identified 100% of the prediction test set to discriminate A. niger from A. ochraceus and A. westerdijkiae; in addition, it had a 86.6% success rate in discriminating A. ochraceus and A. westerdijkiae. This is the first time a study assessed the ability of FT-IR to differentiate A. niger, A. ochraceus, and A. westerdijkiae, and we believe this technique is very promising for classifying and distinguish fungi isolates.