The GARD platform is based on gene expression analysis of stimulated in vitro models of Dendritic cells. By mimicing the initial events of the immune system in response to foreign molecular structures, SenzaGen builds predictive models that are simple, yet able to reflect a complex reality.

Upon exposure, organs and cells in the human body react in a multitude of ways in response to chemicals, proteins or other molecular structures that are recognized as foreign by the immune system. The GARD platform monitors these different reactions by examining transcriptional patterns induced by various stimuli, by the use of predictive genomic biomarker signatures. By tailoring different biomarker signatures to specific biological endpoints of interest, the GARD platform is able to predict adverse effects induced by tested substances with the help of advanced pattern recognition technologies.

How does it work?

GARD is based on prediction models that have “learned” to recognize chemical sensitizers, based on the induced transcriptional profile in the in vitro model of Dendritic cells. The key to GARD’s superior predictive performance lies in the way data are analyzed, and the amount of data generated to make predictions.

By utilizing advanced machine learning techniques, we are able to turn a vast collection of raw data—originating from the measurements of hundreds of genomic biomarkers—into a single answer: An accurate and robust prediction of the tested substance’s adverse effects.