This paper evaluates currently existing computer-based (‘in silico’) models that relate the molecular structure of low molecular weight drugs to their respiratory sensitization danger. The current study focuses on the two primary applications of such structure-activity relationship (SAR) models: disease mechanism theories and toxicological prediction. Analyses of the chemical structures of low molecular weight organic chemicals known to induce occupational asthma have resulted in the creation of mechanistic warnings, which are often based on electrophilic reaction chemistry and protein cross-linking potential. A common feature of substances that have induced human cases of pneumonitis hypersensitivity has also been shown to have protein cross-link potential. Step-by-step modelling of quantitative SAR (QSAR) shows significant increases in prediction of occupational asthma hazards and practical possibilities. There has also been strong evidence for the possible application in predicting toxicology of structural alert mechanistic SARs.

In-vitro and in-silico methods have provided a better knowledge of the molecular interactions between chemical respiratory sensitizers and human protein components.