Repository for PHUSE CSS 2020 hackathon
Welcome to PHUSE CSS Hackathon 2020! The hackathon is taking place:
- Monday 21st September at 13:30-17:30 (EDT)
- Tuesday 22nd September at 09:00-13:00 (EDT)
- Wednesday 23rd September at 13:30-17:00 (EDT)
We will be using the RStudio RTT system, allowing us to use RStudio Pro Products. You will be contacted by the Hackathon leads with information on how to connect. Reach out to the leads if you don't receive this correspondence.
To participate in the Hackathon, make sure that:
- You've created a GitHub account and sent your username to Mike.Stackhouse@atorusresearch.com
- You should have received database credentials for PHUSE Open Data Repository (PODR)
- You're ready to learn and find some new industry solutions!
Many clinical data scientists are expanding their toolset to include more open source solutions like R and Python. This Hackathon will demonstrate the capabilities of both industry-established tools and open source technologies in analysis dataset creation as well the creation of analytics (tables, figures and visualizations).
There will be use cases to demonstrate an overall approach to an open source validation framework. This will be an oral presentation on validation concepts for advanced users The outputs from this Hackathon will be shared with the PHUSE community to be used as a resource by individuals and companies to make the case for expanding the toolset available to clinical data scientists.
The Hackathon consist of three separate streams. These three streams will spend the three days of the Hackathon on their respective tasks, and come together at the end to present their progress. Participants can form different teams within a stream.
This stream will be dedicated to addressing the concerns of environment qualification when using open source languages for analysis. For languages like R, it is largely accepted that the base language itself meets 21 CFR Part 11 compliance, that the installation is qualified and the system itself is qualified. That said, part of the strength of open source languages are the vast third- party contributions available through add-on packages. This stream will work on addressing these issues and providing recommendations on how open source languages can be used safely in regulatory environments.
SDTM and ADaM programming is traditionally done in SAS, and given the extensive use of SAS in regulatory submissions and preparing CDISC data, a large deal of literature exists on how to approach CDISC challenges using SAS. This stream will develop code that demonstrates the production of ADaM datasets using R and Python. Specifications and synthetic data will be provided as uses cases. The data used will be hosted in the PHUSE Open Data Repository (PODR).
This stream will be broken down into two scenarios:
- Low Complexity Datasets
- ADAE
- ADSL
- High Complexity Datasets
- ADLB
- ADVS
The last stream will focus on TLF generation using R and Python. Both of these languages have rich packages available to produce tables and graphs – but clinical reports tend to have standard requirements and nuance that may make this challenging.
The outputs produced by this stream will be motivated by the PHUSE Standard Analyses White Papers, which can be found here. The PHUSE Test Data Factory CDISC Pilot replication data will be used and accessed via PODR.
The targets of this stream will also be broken into low and high complexity scenarios.
- Low Complexity Outputs
- Listings
- High Complexity Outputs
- Adverse Events tables
- Treatment Emergent Adverse Events Risk Difference Plots
- Driven by standard analysis white paper guidance
The PHUSE Hackathon is dedicated to providing a harassment-free conference experience for everyone, regardless of gender, gender identity and expression, sexual orientation, disability, physical appearance, body size, race, age, or level of experience. We do not tolerate harassment of conference participants in any form. If a participant engages in harassing behavior, event organizers retain the right to take any actions to keep the event a welcoming environment for all participants. This includes warning the offender or expulsion from the conference. The Hackathon team is jointly and severally responsible for enforcement of the anti-harassment policy.