Introduction
I returned from Posit Conf 2024 and wanted to reflect on the experience. ProCogia had a really good showing this year and I was happy that I was able to contribute. On Sunday I was able to attend the R-Pharma followed by an all-day workshop on the Pharmaverse. Our very own Lovekumar Patel gave an excellent talk on a custom R package he developed for our clients.
On Wednesday I led a lunch and learn workshop that demonstrated how AI tools can be incorporated into each stage of a data science workflow. ProCogia had a booth at the conference and most of our leadership attended, providing us ample opportunity to meet some of our clients and hear the latest in data science.
R/Pharma
The R-Pharma Roundtable was a full day event that included some fascinating talks about the pharmaceutical industry’s ongoing adoption of R for clinical analysis and reporting. Supporting the use of R in clinical trials is a top priority for ProCogia and we are active across multiple R Consortium Working Groups that are directly supporting this endeavor. My group has successfully migrated several clients from SAS to R but our pharma clients have traditionally been slow to make big changes to their analytics.
This year, something seems to have changed in the market and companies were much more motivated to discuss R for their clinical research and submissions. I was encouraged to see that the conference devoted an entire section open-source initiatives in pharma, including an excellent discussion on effective change management strategies for SAS to R/Python transitions. We have integrated the Altair SAS Language Compiler (SLC) with the new Positron IDE and the potential for this tool to accelerate the transition to R was immediately apparent to attendees.
Pharmaverse
SAS has been the dominant player in the highly-regulated world of clinical research for decades. Although FDA claims to be platform agnostic, the reality is that many of the CDISC standards required for submissions are based in SAS standards. Datasets must be submitted in a .xpt SAS transport files and meta data reported via PROC CONTENTS. Study results are reported using macros developed in SAS. The industry is very invested in SAS and a successful transition to open-source analytics needs to take these existing standards into account.
The Pharmaverse project is a collection of R packages that were developed by a network of companies that wanted to leverage the power of R while still meeting the CDISC requirements for submissions. Development is ongoing, but packages are available for SDTM/ADaM dataset generation, variable derivation, metadata documentation, tabling of results, and more. The project is a great example of how the R community can come together to solve a common problem and I will be incorporating these tools in our SAS to R migration efforts.
Advanced Portfolio Analysis using Shiny
One of the highlights of the conference was a presentation led by one of ProCogia’s stellar consultants. Lovekumar Patel leads our Shiny development team and has developed a custom portfolio management tool for a financial services company. The tool allows the company to track the performance of their investments in real-time and generate reports for their clients. The tool is built in R and uses Shiny to create an interactive dashboard that can be accessed by the company’s clients.
Central to the success of the tool is a novel R package developed by Lovekumar’s team based on the Ag-Grid javascript library. This package addresses deficiencies the team found in existing tabling packages like rhandsontable and dt. Lovekumar received a lot of attention after his talk and there is interest across from other financial services companies in retaining his services for similar projects.
AI tools in data science workflows
On Wednesday, ProCogia hosted a special Lunch and Learn session on the use of AI tools in data science workflows. About a year ago our CEO challenged the company to embrace a culture of AI. The goal was to crowd source the evaluation of the flood of AI tools available on the market with the hope that the cream would rise to the top. We have learned a lot over the past year and we’ve accumulated a variety of simple ways that AI can help us meet our clients’ needs faster and more efficiently.
Essentially, we have found that AI tools can be understood as a talented graduate student that is utterly devoid of common sense. But like any graduate student, it can relieve some of the tedium of your work. Our presentation highlighted the ways that AI tools can be incorporated into each stage of a data science workflow. We demonstrated internal tools that we have developed including SentimentIQ for bringing order to unstructured data, a custom SAS to R translation tool, and incorporating GitHub Copilot and other AI tools directly in Rstudio.
Conclusion
Posit Conf 2024 was a great success for ProCogia and I am proud to have been a part of it. I am looking forward to next year’s conference and I am excited to see how the industry continues to evolve. I am particularly excited to observe the nascent sea change in the pharma industry moving away from SAS and towards open-source analytics.
Explore Our Open-Source Solutions
At ProCogia, we’re not just participants in the open-source community—we’re leaders. Our expertise in R, Python, and advanced data solutions can help your business stay ahead of the curve. Whether you’re looking to migrate from SAS to R, harness the power of Shiny, or develop custom packages, our team has the skills and experience to deliver cutting-edge solutions tailored to your needs.
Learn more about our open-source services and discover how we can drive your next project forward.
