Schedule

Presentation 1

Time: 12:00 - 12:30

Title: High-Throughput Topographical Screening Platforms for Regenerative Medicine

Speaker: Steven Vermeulen

Abstract: At the MERLN Institute, we aim to develop novel technologies to repair tissues and organs. An approach for achieving this is through applying surface structures to control cell behavior. It is known that changing the physical environment can have profound effects on stem cell differentiation or the maintenance of cellular identity. However, it is challenging to identify a structure most suited for a specific application due to the enormous structural design space. Therefore, using a design algorithm, we have generated numerous different patterns, which can first be reproduced on a silicon mold and then imprinted onto polymers using microfabrication. This gave rise to the Topochip platform with 2176 unique micro-topographies and the nanoTopochip with 1249 nano-topographies, allowing us to learn more about the relationship between surface topography and cell response.

Here, I will present a workflow of how we handle such a screen where we process hundreds of thousands of images. I will discuss the application of the software CellProfiler to process these images and machine learning algorithms to associate surface structural design with cell phenotype. Furthermore, I will discuss our ongoing efforts to reach out to the UMC community for novel collaboration opportunities to tackle the challenges associated with big data.

Presentation 2

Time: 12:30 - 13:00

Title: An Ontology-Based Data Integration pipeline for Biomaterials Development for Bone Regeneration

Speaker: Yousra Alaoui Selsouli

Abstract: Calcium phosphate (CaP) ceramics are among the most widely used synthetic bone graft substitutes in the clinic. Nevertheless, potential of most CaPs for inducing bone regeneration is not as high as that of natural bone grafts, and therefore, further improvement is required. Microfluidic technology offers attractive tools for improving and accelerating the development of new materials with higher throughput or in a more controlled manner than the conventional production methods. Using such technologies results in generating large datasets, containing many interrelated parameters involved in the complex mechanism of tissue regeneration. These datasets are extensive, diverse, and often managed independently from one another. This complicates the process of inferring, and validating the role of biomaterials properties on the biological mechanisms involved in biomaterial-induced tissue regeneration. To fully exploit and generate knowledge from these datasets, the specific domain requires a formal description where pertinent concepts, properties, and relationships between those concepts are expressed.

Therefore, our approach consists of reusing existing ontologies related to each field involved in the bone regeneration process, extending the ontologies to fit our requirements, and finally establishing a FAIR pipeline to help with integrating data sources with the selected ontologies using existing programs and services.  The output of this pipeline will allow building a knowledge graph database that can be used to compare experimental data and draw better-informed conclusions. This database will be initially accessible to researchers at MERLN institute and selected collaborating institutes to be queried using elements of the ontologies as predicates, prior to initiating an open access database.

Here, I will present our efforts towards establishing an ontology-based data integration workflow that aims to aid with handling, and mining the large and heterogeneous data sources obtained from our high-throughput microfluidics approach, for a controlled development of biomaterials for bone repair.