Status
Ready
Created On
Updated On
The rapid growth of single-cell transcriptomic technology has produced an increasing number of datasets for both embryonic development and in vitro pluripotent stem cell derived models. This avalanche of data about pluripotency and the process of lineage specification has meant it has become increasingly difficult to define specific cell types or states and compare these to in vitro differentiation. Here we utilize a set of deep learning (DL) tools to integrate and classify multiple datasets. This allows for the definition of both mouse and human embryo cell types, lineages and states, thereby maximising the information one can garner from these precious experimental resources. Our approaches are built on recent initiatives for large scale human organ atlases, but here we focus on the difficult to obtain and process material that spans early mouse, and in particular, human development. Using publicly available data for these stages, we test different deep learning approaches and develop both a model to classify cell types in an unbiased fashion and define the set of genes required to identify lineages, cell types and states. We have used our predictions to probe pluripotent stem cell models for both mouse and human development, showcasing the importance of this resource as a dynamic reference for early embryogenesis.

More details

Use Cases Limitations Evidence Owner's Insight

This app provides a comprehensive platform for exploring gene expression and SHAP (SHapley Additive exPlanations) features in preimplantation mouse and human development. Users can analyze differentially expressed genes (DEGs) across cell types and developmental stages, as well as examine SHAP features to understand the impact of individual genes on model predictions.

For the analysis of differentially expressed genes, users can select the species (human or mouse) and the dataset type (cell type or stage) from a sidebar menu. The app offers options to filter DEGs based on criteria such as log2 fold-change and adjusted p-value, allowing researchers to focus on genes of interest. The resulting subset of DEGs is displayed in a dataframe, facilitating further exploration and analysis.

Additionally, users can explore SHAP features by selecting the dataset (human or mouse) from the sidebar. They can then choose a specific cell type and select genes of interest to examine the SHAP values associated with those genes. The app provides an interactive dataframe displaying the SHAP features, enabling users to identify genes that contribute significantly to model predictions across different cell types.

Overall, this app offers a user-friendly interface for investigating gene expression patterns and understanding the contributions of individual genes to preimplantation development in both mouse and human embryos.

See owner's GitHub repository for more information: https://github.com/brickmanlab/preimplantation-portal.

This app is based on the paper:

Proks, M., Salehin, N., & Brickman, J. M. (2024). Deep Learning Based Models for Preimplantation Mouse and Human Development. bioRxiv, 2024.02.16.580649. https://doi.org/10.1101/2024.02.16.580649

AI models:

Trained models with parameters were uploaded to Hugging Face: https://huggingface.co/brickmanlab/preimplantation-models

Models: scANVI, scANVI [ns=15], XGBoost, scVI, scANVI, scGEN

This application was not uploaded by the author, but through their publicly available Github repository, https://github.com/brickmanlab/preimplantation-portal.

Prototype

Warning: Not intended for clinical use. Assume outputs are unsafe and unvalidated. Use carefully.


  • Favorites: 0
  • Executions: 40

  • Precision Medicine & Genomics

Owner

C Community Discovery

Member since