The R language is rapidly growing into the premier statistical language of choice. Many commercial companies are harnessing the power of R in order to drive business decisions. However, the open source nature and flexibility of the R language, leads to a number of challenges for large organisations.
In particular, many commercial organisations (particularly those in the heavily regulated pharmaceutical and financial industries) demand higher levels of software validation in connection with the use of the open source software.
Due to our extensive knowledge and understanding of the strict regulations involved across a variety of industries, Mango Solutions are able to provide compliant life cycle support to both business and IT functional areas for the validation of R.
Mango have been providing R Services to commercial companies for many years, backed by a full set of quality procedures that ensure validation standards are maintained. We have built on this expertise in order to design and implement a formal R development platform, including the ability to automatically generate tested builds of R.
ValidR is a product we have developed which delivers a validated version of the R language into regulated industries. It is designed to transform a language that, as a standard installation provides “absolutely no warranty”, into a system whose scripts comply with FDA guidelines on the qualification and validation of systems (21 CFR Part 11).
With one of the largest R consultancy teams in the world and a long history of delivering R consultancy and software products within regulated industries Mango is uniquely positioned to create the ValidR product. The Mango business and its ISO9001 accredited Quality Management System (QMS) are subject to regular external audits that have helped shape the R validation process.
Mango Solutions’ performance qualification provides risk-based documented evidence that the R environment functions identically to the standard version of the R application with the addition of;
- additional unit testing,
- a customised installation platform with automated tools to build R core source and supporting packages;
- version controlled repositories to support software practices;
- security enhancements required by the customer to restrict end users’ ability to modify the functionality of R application and environment.
The purpose of the R environment is for end users to create and execute scripts written in the R language that perform data dependent statistical analysis for which there exist business requirements.
Due to the complex nature and end user approach to using statistical reporting software which requires the development of scripts, a risk-based approach to validation must be taken. The validation activities for performance qualification must ensure that the delivered code is as intended and produces expected results when independent units or components of the code perform tasks collectively.
When installing an R environment in a GCP environment, performance qualification procedures follow the validation requirements for one-off programs.
One-off programs are programs or scripts, including macros, written for use with one set of data e.g. an analysis on a single study.
As with any software for use in a regulated environment, validation procedures of one-off programs should reflect the size and complexity of the one-off program, the extent of its intended use and the potential consequences of any error.
The Performance Qualification
The performance qualification will be run and documented within a test environment. Upon successful completion of the performance qualification within the test environment, the R environment can be deployed within the production environment.