In the present study, Raman spectroscopy is employed to assess the potential toxicity of chemical substances. Having several advantages compared to other traditional methods, Raman spectroscopy is an ideal solution for investigating cells in their natural environment. In the present work, we combine the power of spectral resolution of Raman with one of the most widely used machine learning techniques. Support vector machines (SVMs) are used in the context of classification on a well established database. The database is constructed on three different classes: healthy cells, Triton X-100 (necrotic death), and etoposide (apoptotic death). SVM classifiers successfully assess the potential effect of the test toxins (Triton X-100, etoposide). The cells that are exposed to heat (45 degrees C) are tested using the classification rules obtained. It is shown that the heat effect results in apoptotic death, which is in agreement with existing literature.