#GenomicsFeatures ###Analysis of Genomics Feature Matrices

This project contains a Google App Engine application written in Python that performs the following tasks:

• Converts an n x m feature matrix in tsv format into a file in csv format with n x m rows of tuples (feature, sample, value) - run_fmx_convert.py
• Builds a model for this data in Datastore - model.py
• Runs a map-only MapReduce job to import the csv file you converted from GCS and into the model - pipeline.py
  • Data can also be imported via the bulkload mechanism but this is very slow. - run_upload_data.sh
• Defines and builds full text search index over the feature names. - search.py
• Provides a User Interface that allows you to do the following: - main.py
  • Perform most of the admin tasks listed above.
  • Find features by using the full text search.
  • Pull up samples and their values for a given feature, including a chart that graphs the distribution of the values.
  • Fetch the value for a given sample and feature.
• Provides a simple REST API that lets you select features and samples and will return their values. main.py
• An example R script that can be used to call the API to fetch feature matrix data and perform analysis on this in R. - Analysis.R

This application can be run in both Google App Engine (Classic) or in a Google Managed VM via Docker.