CBBS Subgroup Contributed Talks

Donggu Lee

Konkuk University

"Role of OCT1 in regulation of miR-451-LKB1-AMPK-OCT1-mTOR core signaling network and cell invasion in glioblastoma"

Glioblastoma multiforme (GBM) is the most aggressive form of brain cancer with a short central survival time. GBM is characterized by aggressive proliferation and critical cellular infiltration. miR-451 and its downstream molecules (LKB1, AMPK, OCT1, mTOR) are known to play a pivotal role in balancing proliferation and aggressive invasion in response to metabolic stress in a tumor microenvironment (TME). Recent studies have shown that OCT1 and LKB1 play an important role in regulation of the mutual inhibition between cell proliferation and migration. In this work, we develop a mathematical model of signaling pathway dynamics in GBM evolution which focuses specifically on the relative balance of proliferative capacity and invasion potential. In the work, we represent the miR-451/LKB1/AMPK/OCT1/mTOR pathway by a mathematical model and show how the effect of fluctuating glucose on tumor cells needs to be reprogrammed by taking into account the recent history of glucose variations and an LKB1-OCT1 mutual feedback loop. The simulations show how changes in glucose have a significant effect on the level of key signaling molecules, determining in promotion or inhibition of glioma cell migration (Kim, Lee, & Lawler, Phil Trans Roy Soc-B, 2020).

Yukitaka Ishimoto

Akita Prefectural University

"In-vivo cell flow visualisation using deep learning and other means"

In recent years, measurements of cellular movements and forces in living body have been paid much attention, chiefly for regenerative therapy and medical applications. It is because they are thought to give deeper insight on tissue mechanics and engineering. There are various ways of invasive and non-invasive measurements. Among them, cell deformations and flow play an pivotal role for elucidation of tissue/organ morphogenesis. However, conventional flow visualization techniques, such as PIV and PTV, often fail to capture the cell flow due to cellular morphological events. To adequately develop such measurements, it is critical to establish precise detection of positions/shapes and correspondence between individual cell shapes at different timepoints. In this work, we show our two distinct attempts of flow visualization of deforming epithelial sheet. One is for particle tracking velocimetry (PTV) of four-dimensional cell flow by using deep neural network model (DNN) onto deforming nucleus images. The other is to track cell shape changes of the sheet by extracting cell boundaries from live-imaging data and further fitting them to a vertex-edge configuration of the bubbly vertex model. Further extensions of both attempts will also be discussed.

Tara Hameed

Imperial College London

"A systematic workflow to assess the useability of data in model development"

Data sparsity is one of the bottlenecks we often encounter in model development, especially for disease modelling or in fields where interdisciplinary cross-collaboration is still being developed. When a model is fit to sparse data, it is hard to discern whether potential model misfit is caused by inherent model misspecification, which requires reformulation of the model, or by data sparsity, which requires further data collection. We proposed a systematic workflow to assess the degree to which the available data can inform mathematical models theoretically, by upcycling a known statistical workflow that uses simulation studies. The proposed workflow quantifies the useability of the experimental data in terms of expected quality of parameter estimation and model prediction. Application of the workflow to our mathematical model of early-stage invasive aspergillosis (pulmonary fungal infection), adapted from a previous model, allowed us to suggest future experiments that could provide more “useable” data to infer the model's nonlinear interaction parameters and to make better predictions. The presented workflow could be useful when models are developed with data sparsity as a limiting factor for model-based inference.

Anibal Thiago Bezerra

Instituto de Ciências Exatas, Universidade Federal de Alfenas

" Gastric Emptying and Orocaecal Transit Analyses of Diabetic and Control Individuals Through Deep Neural Networks"

Classical analysis of experimental data generally relies on statistical methods. These methods, however, can be contradictory depending on the methodology and the adopted metrics. In the quantification of gastric emptying (GE) and orocaecal transit (OCT), this is the case in the discrimination between rats who have dysfunctions or diseases like diabetes and the ones that do not. Metrics involved in this context are mean gastric emptying time (MGET), orocaecal transit time (OCTT), and mean caecum arrival time (MCAT). To overcome their limitations, here we present an artificial neural network (ANN) capable of discriminating between control and diabetic individuals rats through GE and OCT data analysis of alternate current biosusceptometry (ACB). For GE, the ANN classification reached an accuracy above 90% after a few epochs. The respective sensitivity was 88%, and the specificity was 83%. For OCT, the accuracy also achieved 90%, with a specificity of 75% and unitary sensitivity. These achieved results support that the proposed ANN can be an alternative methodology to the classical method employed over the years in the gastrointestinal transit area. This work is supported by funding from grant #2020/05556-0, São Paulo Research Foundation (FAPESP).