Fast Development and Robust Deployment of Data Driven Bioinformatics Web Apps and Workflows

About this Class

By Avi Ma'ayan, Director of the Mount Sinai Center for Bioinformatics Daniel Clarke, Senior Data Scientist and Nicole Moiseyev, Summer Scholar Department of Pharmacological Sciences, Mount Sinai Center for Bioinformatics, Icahn School of Medicine at Mount Sinai, New York, NY In this presentation we will discuss how we are transitioning from hosting the web-based bioinformatics applications we develop from a Mesos-Marathon cluster environment to a Rancher/Kubernetes environment. We are also moving away from hosting full stack web-based bioinformatics applications such as Enrichr, Harmonizome, BioJupies and ARCHS4 towards developing appyters. Appyters turn Jupyter notebooks into fully functional standalone web applications. They extend the concepts developed for BioJupies to many other applications. Appyters present to users a data entry form that enables them to upload their data and set various parameters for a multitude of bioinformatics analysis pipelines. Once the form is filled, the Appyter executes the corresponding notebook online, saving the output without having to interact directly with the code. Appyters can be applied to a variety of workflows including building customized machine learning pipelines, analyzing RNA-seq data, and producing publishable figures. Appyters enable the rapid development of web-based applications as demonstrated by the breadth of over 50 examples. Join Webex meeting