Biology subjects

Wierling, C.

Publications and source records attributed to Wierling, C..

3 recordsLinked to original sources

bioRxiv2021

Isolation, omics characterization and organotypic culture of alveolar type II pulmonary epithelial cells

The alveolar type II (AT2) epithelial cell fraction includes the stem cells of the pulmonary alveoli, functioning in lung homeostasis and post-injury repair. AT2 cells have been characterized primarily in situ, in transgenic mouse models. We report a new methodology for their isolation, their "omics" characterization and stroma-cell-free organotypic culture. Our multi-omics analysis identified high expression of genes involved in oxidative phosphorylation and of AP-1 components, as well as new phosphorylation sites in AT2 biomarkers. Furthermore, we show that supplementation with KGF, FGF10 & HGF suffices for the in vitro proliferation of AT2 cells and formation of alveolar organoids, suggesting that AT2-based organotypic development depends on ligands of the c-Met and FGFR2 receptors. The reported methodology and in-depth molecular characterization provide new tools for the in vitro and in vivo functional analysis of pulmonary cells and of mouse models of lung disease.

cell biology↗

bioRxiv2019

Mini-batch optimization enables training of ODE models on large-scale datasets

Quantitative dynamical models are widely used to study cellular signal processing. A critical step in modeling is the estimation of unknown model parameters from experimental data. As model sizes and datasets are steadily growing, established parameter optimization approaches for mechanistic models become computationally extremely challenging. However, mini-batch optimization methods, as employed in deep learning, have better scaling properties. In this work, we adapt, apply, and benchmark mini-batch optimization for ordinary differential equation (ODE) models thereby establishing a direct link between dynamic modeling and machine learning. On our main application example, a large-scale model of cancer signaling, we benchmark mini-batch optimization against established methods, achieving better optimization results and reducing computation by more than an order of magnitude. We expect that our work will serve as a first step towards mini-batch optimization tailored to ODE models and enable modeling of even larger and more complex systems than what is currently possible.

systems biology↗

bioRxiv2017

Efficient parameterization of large-scale mechanistic models enables drug response prediction for cancer cell lines

The response of cancer cells to drugs is determined by various factors, including the cells mutations and gene expression levels. These factors can be assessed using next-generation sequencing. Their integration with vast prior knowledge on signaling pathways is, however, limited by the availability of mathematical models and scalable computational methods. Here, we present a computational framework for the parameterization of large-scale mechanistic models and its application to the prediction of drug response of cancer cell lines from exome and transcriptome sequencing data. With this framework, we parameterized a mechanistic model describing major cancer-associated signaling pathways (>1200 species and >2600 reactions) using drug response data. For the parameterized mechanistic model, we found a prediction accuracy, which exceeds that of the considered statistical approaches. Our results demonstrate for the first time the massive integration of heterogeneous datasets using large-scale mechanistic models, and how these models facilitate individualized predictions of drug response. We anticipate our parameterized model to be a starting point for the development of more comprehensive, curated models of signaling pathways, accounting for additional pathways and drugs.

systems biology↗