%0 Journal Article %1 dirk2022recognition %A Dirk, Robin %A Fischer, Jonas L. %A Schardt, Simon %A Ankenbrand, Markus J. %A Fischer, Sabine C. %D 2022 %J arXiv %K cctb csi myown sabinefischer %P 2212.10058 %T Recognition and reconstruction of cell differentiation patterns with deep learning %U http://arxiv.org/abs/2212.10058 %X Cell lineage decisions occur in three-dimensional spatial patterns that are difficult to identify by eye. There is an ongoing effort to replicate such patterns using mathematical modeling. One approach uses long ranging cell-cell communication to replicate common spatial arrangements like checkerboard and engulfing patterns. In this model, the cell-cell communication has been implemented as a signal that disperses throughout the tissue. On the other hand, machine learning models have been developed for pattern recognition and pattern reconstruction tasks. We combined synthetic data generated by the mathematical model with deep learning algorithms to recognize and reconstruct spatial cell fate patterns in organoids of mouse embryonic stem cells. A graph neural network was developed and trained on synthetic data from the model. Application to in vitro data predicted a low signal dispersion value. To test this result, we implemented a multilayer perceptron for the prediction of a given cell fate based on the fates of the neighboring cells. The results show a 70% accuracy of cell fate reconstruction based on the nine nearest neighbors of a cell. Overall, our approach combines deep learning with mathematical modeling to link cell fate patterns with potential underlying mechanisms.

@article{dirk2022recognition, abstract = {Cell lineage decisions occur in three-dimensional spatial patterns that are difficult to identify by eye. There is an ongoing effort to replicate such patterns using mathematical modeling. One approach uses long ranging cell-cell communication to replicate common spatial arrangements like checkerboard and engulfing patterns. In this model, the cell-cell communication has been implemented as a signal that disperses throughout the tissue. On the other hand, machine learning models have been developed for pattern recognition and pattern reconstruction tasks. We combined synthetic data generated by the mathematical model with deep learning algorithms to recognize and reconstruct spatial cell fate patterns in organoids of mouse embryonic stem cells. A graph neural network was developed and trained on synthetic data from the model. Application to in vitro data predicted a low signal dispersion value. To test this result, we implemented a multilayer perceptron for the prediction of a given cell fate based on the fates of the neighboring cells. The results show a 70% accuracy of cell fate reconstruction based on the nine nearest neighbors of a cell. Overall, our approach combines deep learning with mathematical modeling to link cell fate patterns with potential underlying mechanisms.}, added-at = {2023-04-28T19:40:01.000+0200}, author = {Dirk, Robin and Fischer, Jonas L. and Schardt, Simon and Ankenbrand, Markus J. and Fischer, Sabine C.}, biburl = {https://www.bibsonomy.org/bibtex/2dad6824c6f062bd40596c32bdefa28dd/scfischer}, description = {[2212.10058] Recognition and reconstruction of cell differentiation patterns with deep learning}, interhash = {8fed3f5d5454e1e80a4614313fde640b}, intrahash = {dad6824c6f062bd40596c32bdefa28dd}, journal = {arXiv}, keywords = {cctb csi myown sabinefischer}, note = {cite arxiv:2212.10058}, pages = {2212.10058}, timestamp = {2023-05-22T14:57:16.000+0200}, title = {Recognition and reconstruction of cell differentiation patterns with deep learning}, url = {http://arxiv.org/abs/2212.10058}, year = 2022 }