Johns Hopkins Precision Medicine Analytics Platform (PMAP) Cookbook

About the PMAP Cookbook

Table of Contents

Before you start cooking

• Do you need an IRB to acccess PMAP
• What type of protocol should I submit for my research use of PMAP?
• Do I have to get Data Trust approval to use PMAP?
• How can I tell what data are available in PMAP?

Intro to Crunchr and Jupyter Notebooks

• Crunchr collaborative data science platform
  • How to access the PMAP Cookbook on Crunchr
  • Steps to using Crunchr for collaborative research
• Introduction to Jupyter notebooks
  • How to access files from your Johns Hopkins H: drive
  • How to access files from your Johns Hopkins S: drive

Accessing Your Data

• Accessing your data through Python
• Accessing your data through RStudio
• Accessing your data with Stata

Working with EMR Data (Python)

• Connecting to the PMAP database server
  • Requesting access to the asthma dataset
• Intro to SQL for mining EMR data
• Ashthma cohort discovery with regular expressions and set operators
• Exploratory data analysis and visualization of EMR data
  • Data cleaning example
• Data quality assessment example
• Statistical testing EMR Asthma example

Conducting NLP Projects (Python)

• Tokenizing a corpus of text
• Matching text with Regular Expression
• ngrams and term weighting
• Document classification

Working with Medical Images (Python)

• How to Access DICOMs From XNAT
• Working with DICOM image files
• Radiomics Example
• High Perfomance Compute (HPC) clusters for Image Analysis
• Hello World Chest X-Ray Deep learning example

Programming Resources

• Python Data Science Handbook (O'Reilly Media)
  • Python programming
  • Data Manipulation with Pandas
  • Data Visualization with Matplotlib
  • Introduction to Numpy
  • Machine Learning with Scikit