Open Source in Pharma with Phil Bowsher

Transcript#

This transcript was generated automatically and may contain errors.

I think the role that I have helped bring to this space is bringing people together and so I think through my work at Posit I saw that so many of these pharmaceutical companies just wanted to get together they wanted to talk and they were coming to me to say can you make these introductions so starting R and Pharma was really the spark that brought a lot of the pharmaceutical companies together and there for the first time the R Validation Hub had their hangout you know ideas around the Pharmaverse were sparked out of the idea that these pharmaceutical companies had specific workflows and code that they wanted to share so it's been amazing to really just see the life that has come out of the R and Pharma conference there was very much this sense and momentum that a new horizon was coming out for open source to change the landscape and lead to a lot of innovations for clinical trials and so a lot of groups that had implemented R for research or for Shiny had now started saying can we use this in addition for our late-stage clinical trials work.

I could certainly feel that there was a change coming to the way that clinical trials were conducted in bringing open source to that ecosystem but I knew that it was going to take some time and there was going to be a decent amount of change management that would have to take place for that to work. Part of the popularity of R in the drug development space especially late-stage clinical trials is because there was a legacy stat programming language that people had used and so migrating from that language to R was a very easy transition for a lot of organizations because R is an open source language built for statistical computing as well as that legacy code and programming tool was also built for stats programming so I think the transition from that previous tool to R is natural because languages like Python are more general purpose programming languages and so I think that has led to a lot of the adoption in the space as well as Shiny I would say.

Part of the popularity of R in the drug development space especially late-stage clinical trials is because there was a legacy stat programming language that people had used and so migrating from that language to R was a very easy transition for a lot of organizations because R is an open source language built for statistical computing.

Posit's professional tools in pharma

A lot of pharmaceutical companies use our professional software for their clinical trials and that typically consists of Posit team and Posit team is made up of three different products there's a Posit workbench which is the development environment literally it's called a development environment because it's where people are going to go in and do their programming. There they create software and artifacts and content and Shiny apps and workflows and then they move that into production into our second product which is called Posit Connect that's the production and deployment platform that we host and maintain and then the back-end what's surfacing out the production and oftentimes validated packages for both of those environments is the Posit package manager and so that's feeding packages for reproducibility into the development side as well as feeding packages into the production side.

Pharmaceutical companies care a lot about Posit package manager because there is this great desire to own and manage the validation and oftentimes what's referred to as qualification of packages and so a lot of work goes internally at pharmaceutical companies to say these are the core packages typically around a hundred to two or three hundred or so packages that this is what people are going to use for clinical trials so package manager specifically is critical for clinical development because it makes the source of where their package is coming from locked down and approved for the work internally at the company.

Yeah there's a lot of use cases and companies and pharmaceutical organizations using our professional tools. Recently this was highlighted by Satish Murthy in the data science hangout that they use Posit package manager for surfacing and managing the source of their validated packages that they bring out. That's one of the most key things for organizations is to make sure the source of the packages coming into the workflows is validated and the right packages. In addition to that Merck also had a webinar with us that highlighted their use of package manager. There's lots of other organizations that have highlighted the use of Posit Connect for where they manage and host their Shiny applications. For example Roche and Genentech in their webinar a couple months ago highlighted that they have an entire framework that they've built in house around Shiny and they host their Shiny apps there with Posit Connect.

It's going to be interesting to see how the future plays out with introductions to new tooling like Python and leading that into AI and how some of that could change some of the clinical trials. I don't think that's going to happen right away but we see this very big inclusive and interoperable future for drug development.