Thorin Tabor - Democratizing the accessibility of computational workflows | JupyterCon 2020

Brief Summary One of the difficulties faced by notebook developers is the handoff from the original author to collaborators at differing levels of technical sophistication. GenePattern Notebook aims to make Jupyter accessible to all. This empowers organizations to best utilize the latest analysis methods and enables researchers to explore available datasets, regardless of their programming abilities. Outline One of the difficulties faced by notebook developers is the handoff from the original author to collaborators at differing levels of technical sophistication. Whether they are students, bench scientists, product owners or management, significant technical hurdles often must be overcome before a notebook’s workflow can be successfully reproduced. GenePattern Notebook aims to make Jupyter accessible to all. It builds upon the new features and capabilities of JupyterLab, allowing notebooks to become a conversation between collaborators, rather than a one-way handoff. This empowers organizations to best utilize the latest analysis methods and enables researchers to explore available datasets, regardless of their programming abilities. To achieve this, GenePattern Notebook provides a suite of extensions that offer a wide range of enhancements, including the capability to render any Python function as an interactive widget, a rich-text editor for markdown cells, a graphical interface for adding new analytic tools to a notebook and the GenePattern Notebook Workspace, a free JupyterHub instance where researchers can develop, run and publish their own bioinformatics notebooks. GenePattern Notebook also allows Jupyter to communicate with the open source GenePattern API. This service wraps hundreds of different software tools for general machine learning methods—including clustering, classification and dimension reduction—as well those in the bioinformatics domain, including tools for analyzing gene expression data, sequence variation, proteomics and genomic networks.