SCAnDi: Single-cell and single molecule analysis for DNA identification

Every nucleated cell in our bodies contains at least one copy of our entire genome, and thus has the potential to identify us. We shed cells wherever we go, often in a context specific manner: different types of contact will result in the transfer and mixture of different cell types. In samples derived from sexual assaults, the cells and DNA recovered will include mixtures of the victim(s) and attacker(s); other contacts including 'touch' DNA samples will contain cells and DNA from multiple contributors. Indeed, up to 45% of forensic samples within the UK criminal justice system contain human DNA from two or more individuals, often preventing successful search of these profiles against DNA databases.

The aim of this project is to combine advances in single-cell isolation and analysis with established and novel approaches for DNA profiling. Typically when DNA is recovered from a mixed sample, cellular material is lysed and cellular and acellular DNA molecules are mixed, resulting in a complete loss of morphological information about the cell the DNA originated from. Obtaining cell-of-origin information could make a critical difference in deconvoluting these samples - especially higher order mixed samples - and ascribing a contextual narrative as to what cell the DNA came from, how it was transferred and by whom.

Single-cell genomics has advanced rapidly over the last decade. We can now measure genetic diversity in individual cells, as well as molecular markers of cell "type" in addition to microscopic analysis of cell morphology. These technological advances have revolutionised studies in developmental and cancer biology but still remain largely unexplored in a forensic context.
Laser Capture Microdissection (LCM) has previously been demonstrated to have applicability in deconvoluting cell mixtures, but advances in molecular biology, genomics, microfluidics and imaging cell sorting now offer the potential to isolate and profile individual cells. In a forensic setting this could enable: 1) obtaining single cells for DNA profiling; 2) the deconvolution of mixed DNA profiles - where DNA / cells from multiple individuals are present in a sample; 3) Linkage of specific cell types with DNA profiles (phenotype or epigenetic/transcriptomic markers); and 4) separation of cellular DNA from background DNA.

Our multidisciplinary team will develop approaches for single-cell separation and subsequent single-cell DNA profiling using conventional and next-generation sequencing approaches, benchmarking the readout against gold standard profiling on bulk material. We will deliver proof-of-principle data from individual human cells, including artificial mixtures of single-source cells as well as post-coital mixtures. We will demonstrate the capability of single-cell profiling to distinguish individual donor information from complex mixtures. Furthermore, we will demonstrate the potential to assign cell type information to single-cell DNA profiles using imaging cell sorting, coupled with automated approaches for cell-type classification. These novel approaches need proof-of-principle data to test and demonstrate their ability to support forensic investigations, but we anticipate the application of this technology will complement traditional methods to considerably improve the reliability of such information as intelligence and as evidence in court.

Critically, we will use this project as a platform for stakeholder engagement and outreach. We will establish a network of forensic practitioners, commercial suppliers, researchers and investigative and legal representatives and host two in-person workshops and additional bi-monthly meetings, to ensure the technical deliveries remain aligned with the needs of end-users. The project will also enable a framework for sample/data acquisition and transfer between our labs, enabling access to critical infrastructure and capability to support sustainable, longer-term community building.