2018 SLAS Advanced 3D Human Models and High-Content Analysis Conference

The possibilities for imaging biological samples are constantly evolving as novel imaging modalities are developed and applied to ever wider areas of research.  With every hardware advancement, researchers are obliged to develop methods to extract, analyse and interpret data from these images as well as re-evaluate the data that can be extracted from existing imaging modalities.  This session will highlight just some of the many current and near-future techniques in both arms of this on-going evolution.

Applying high content 3D image analysis to study non-alcoholic fatty liver disease progression
Fabián Segovia-Miranda 
Histological analysis of human liver biopsies represents the gold standard to morphologically characterize tissue alterations induced by non-alcoholic liver disease (NAFLD). However, this approach is semi-quantitative, subjective and lacks three-dimensional (3D) information, e.g. structure of the bile canaliculi network. We developed a flexible pipeline to reconstruct 3D digital models of tissue from high-resolution multiphoton microscopy images. We then used these models for high-content analysis of liver tissue at different stages of NAFLD progression. We found profound geometrical and topological defects in the bile canaliculi network which correlates with disease progression. Moreover, we identified several potential tissue biomarkers (e.g. zonated lipid droplet accumulation, nucleus homogeneity). This strategy allows for a systems biology approach to connect the microanatomy of bile canaliculi with biliary fluid dynamics within the lobule. Generation of geometrical models of human tissues offers an unprecedented way of analyzing pathophysiological processes, revealing unknown mechanisms of disease establishment and progression.

Machine Learning Enables Live Label-Free Phenotypic Screening in Three Dimensions 
Eoghan O'Duibhir
There is a large amount of information in brightfield images that was previously inaccessible by using traditional image analysis techniques. This information can now be exploited by using machine-learning approaches for both image segmentation and the classification of objects. We have combined these approaches with a label-free assay for growth and differentiation of leukemic colonies, to generate a novel platform for phenotypic drug discovery. Initially, a supervised machine-learning algorithm was used to identify in-focus colonies growing in a three-dimensional (3D) methylcellulose gel. Once identified, unsupervised clustering and principle component analysis of texture-based phenotypic profiles were applied to group similar phenotypes. In a proof-of-concept study, we successfully identified a novel phenotype induced by a compound that is currently in clinical trials for the treatment of leukemia. These methods are now also being used for the screening of both zebrafish and human single cell derived liver organoids at our facility. We believe that this approach will be of great benefit for the utilization of patient-derived 3D cell culture systems for both drug discovery and diagnostic applications.