EPFL Platform Technical Advancement Seed Fund

Aye, Yimon (Laboratory of Electrophiles and Genome Operation, LEAGO)

Pavlou, Maria (Proteomics Core Facility, PTP)

The omics/sequencing tools are of limited ability to inform on locale-specific actionability by small-molecules, restricting fundamental scientific advancement as well as drug development.

The aim of the project is to fill this gap by offering new ways to probe and prove on-target interactions directly in vivo, bringing otherwise-overlooked functional contexts into the latest spatial-omics tools. Central to this project is the goal to address current challenges at EPFL proteomics core facility, from multiplex-data deconvolution to streamlining experimental design.

Compatibility of so-developed workflow will be road-tested, using samples stemming from applying the teams’ own functional- and spatial-omics tools in multiple live models: cells, worms, and fish. Direct benefit is evidenced by 20 different SV/STI/SB-labs supporting this project. Through this work, it is expected to gain unmet opportunities to establish user-friendly protocols that promise to become world-class go-to workflows for the broader EPFL communities.   

Michaud, Véronique (Laboratory for Processing of Advanced Composites, LPAC)

Turberg, Pascal (ENAC Interdisciplinary Platform for X-ray micro-tomography, PIXE)

Partners:

Grossiord, Charlotte (Plant Ecology Research Laboratory, PERL)

Vassilopoulos Anastasios (Composite Construction Laboratory, CCLAB)

The objective of this project is to increase the potential of the PIXE X-Ray tomography platform installed at EPFL ENAC by co-developing the EVITA device, a unique prototype developed and built at CSEM for X-Ray phase contrast radiography and tomography of large-scale samples, using Talbot Lau interferometry, which is now installed at EPFL within the LPAC (STI). The strength of this phase contrast imaging technique is that it brings out with clarity the presence and distribution of phases, which are usually difficult to differentiate by X-Ray absorption, including heterogeneous fibrous media of similar density, such as organic materials, or carbon reinforced composite materials.

The objectives are to start a collaboration between LPAC and PIXE to maintain and improve the current EVITA software for acquisition and reconstruction of images; develop and install testing rigs for in-situ tests within the equipment, including heating or humidity controlled tests; broaden the technique currently used at LPAC for dynamic flow imaging of composite processes towards in-situ composite mechanical tests, as well as observation of living plants, biological or other heterogeneous materials to initiate further projects and collaborations within EPFL and beyond.

Pardon, Gaspard (Bioengineering & Technology Platform, PTBET)

Huelsken, Joerg (Huelsken Lab, UPHUELSKEN)

Merten, Christoph (Laboratory for Biomedical Microfluidics, LBMM)

Oricchio, Elisa (Elisa Orrichio’s Lab, UPORICCHIO)

The recent emergence of organoids as in vitro models, as well as the use of fresh tissue explants, for in vitro testing hold great promises towards the development and translation of novel therapies and of personalised medicines. However, the deployment and scaling up of such in vitro human models in translational research and applications requires expertise aggregation and a dedicated infrastructure with custom-made automation capabilities.

The aim of the project is to develop automation protocols on a robotic platform, as well as custom microfluidic components, for the culture, handling and assay of such 3D in vitro models, including organoids and microscopic tissue explant. Furthermore, the goal is to develop smart imaging algorithms for intelligent culture optimization. The results of this project will be made available to the EPFL community and beyond through the BET platform and will serve as foundation for the future development and acquisition of a fully integrated automation platform. As such, it will pave the way towards the establishment of a ‘live biobank’ of patient-derived organoids.

Petersen, Carl (Sensory Processing Laboratory, LSENS)

Schneider, Bernard (Bertarelli Foundation Gene Therapy Core Facility, PTBTG)

Over the last decade, rapid progress has been made in generating genetic tools to optically measure and manipulate neuronal activity. One of the remaining bottlenecks is the need for effective technologies to express sensors and optogenetic tools over large brain regions. So far, this has typically required complex and time-consuming mouse genetic engineering and breeding strategies. Recently, a new gene delivery approach has emerged using adeno-associated viral (AAV) vectors engineered to cross the blood-brain barrier. These vectors need to be optimized to drive expression of the proteins at levels, which are compatible with optical recording or manipulation of neuronal activity, while avoiding the toxic effects often encountered with chronic high-level expression.

The project aims to establish adeno-associated viral (AAV) vectors technology for the monitoring and manipulation of neuronal activity at large scale. The Bertarelli Foundation Gene Therapy Platform (PTBTG) led by Bernard Schneider will develop and produce a new set of vectors for imaging neuronal activity (genetically encoded fluorescent indicators) and manipulating neural circuit function (optogenetic actuators). The Laboratory of Sensory Processing (LSENS) led by Carl Petersen will test and apply the new vectors for measurement and manipulation of neural circuit function in the mouse brain using state-of-the-art optical technologies.

Mensi, Mounir (X-Ray Diffraction and Surface Analytics Platform, ISIC-XRDSAP)

Smit, Berend (Laboratory of Molecular Simulation, LSMO)

Marzari, Nicola (Laboratory of Theory and Simulation of Materials, THEOS)        

X-Ray photoelectron spectroscopy (XPS) is a surface-sensitive technique that found mounting use for the characterisation of increasingly complex materials. However, data analysis tools did not keep up with the recent developments, and to date, the interpretation of photoelectron spectra mostly relies on the experience of the researchers, and can therefore be prone to incorrect assignments. This might seem surprising given the recent developments in the simulation of photoelectron spectra (many led by co-applicants of this application), which makes it possible to predict highly accurate spectra that enables data-driven, objective, interpretation of the data. These developments, however, did not propagate to experimentalists due to technical barriers in setting up the simulations.

In this project, the aim is to remove these technical barriers by making simulated spectra directly accessible from an electronic lab notebook, in which platform users analyse their experimental spectra. To also allow for the analysis of complex systems, the approach will be augmented with a machine-learning model trained on fingerprint spectra.

Mensi, Mounir (X-Ray Diffraction and Surface Analytics Platform, ISIC-XRDSAP)

Smit, Berend (Laboratory of Molecular Simulation, LSMO)

Marzari, Nicola (Laboratory of Theory and Simulation of Materials, THEOS)        

X-Ray photoelectron spectroscopy (XPS) is a surface-sensitive technique that found mounting use for the characterisation of increasingly complex materials. However, data analysis tools did not keep up with the recent developments, and to date, the interpretation of photoelectron spectra mostly relies on the experience of the researchers, and can therefore be prone to incorrect assignments. This might seem surprising given the recent developments in the simulation of photoelectron spectra (many led by co-applicants of this application), which makes it possible to predict highly accurate spectra that enables data-driven, objective, interpretation of the data. These developments, however, did not propagate to experimentalists due to technical barriers in setting up the simulations.

In this project, the aim is to remove these technical barriers by making simulated spectra directly accessible from an electronic lab notebook, in which platform users analyse their experimental spectra. To also allow for the analysis of complex systems, the approach will be augmented with a machine-learning model trained on fingerprint spectra.