GCxGC-TOFMS and supervised multivariate approaches to study human cadaveric decomposition olfactive signatures

In forensic thanato-chemistry, the understanding ofthe process of soft tissue decomposition is still limited. Abetter understanding of the decomposition process and thecharacterization of the associated volatile organic compounds(VOC) can help to improve the training of victim recovery(VR) canines, which are used to search for trapped victimsin natural disasters or to locate corpses during criminal investigations.The complexity of matrices and the dynamic natureof this process require the use of comprehensive analyticalmethods for investigation. Moreover, the variability of theenvironment and between individuals creates additional difficultiesin terms of normalization. The resolution of the complexmixture of VOCs emitted by a decaying corpse can beimproved using comprehensive two-dimensional gas chromatography(GC× GC), compared to classical singledimensionalgas chromatography (1DGC). This study combinesthe analytical advantages of GC×GC coupled to timeof-flight mass spectrometry (TOFMS) with the data handlingrobustness of supervised multivariate statistics to investigatethe VOC profile of human remains during early stages ofdecomposition. Various supervised multivariate approachesare compared to interpret the large data set. Moreover, earlydecomposition stages of pig carcasses (typically used as humansurrogates in field studies) are also monitored to obtain adirect comparison of the two VOC profiles and estimate therobustness of this human decomposition analog model. In thisresearch, we demonstrate that pig and human decompositionprocesses can be described by the same trends for the majorcompounds produced during the early stages of soft tissuedecomposition.