PHUSE EU Connect 2021 - six abstracts accepted!

The PHUSE EU Connect 2021 will be a virtual event again, taking place on 15th–19th November. That means that we can’t meet in person, but OCS Life Sciences will actively participate with five presentations and one poster!  We are pleased to announce that OCS Life Sciences will deliver a contribution on the following topics:  

Caro Sluijter - Senior Statistical Programmer - OCS Life Sciences
"Automation of Process Flow Instances"

One of the notable features of SAS® Life Science Analytics Framework (LSAF) is the workflow capability. This functionality allows you to automate processes by using process flows; to allocate tasks and to keep track of the progress of tasks and activities. A process flow is particularly useful when a process consists of several standardised tasks that need to be executed repetitively within or across studies. Process flows can thus eliminate repetitive tasks.
However, setting up process-flow instances can in itself be a repetitive task, because for all process flow elements the details need to be selected manually. As process flows are meant to remove repetitiveness, we also wanted to remove repetitiveness in creating the process flow instance.
In this paper we will demonstrate how LSAF API macros can be used to set up a process flow instance requiring as little as one or two parameters.

Louella Schoemacher - Senior Statistical Programmer - OCS Life Sciences

"Utilising LSAF for data management at a Belgian biotech"

While PC-SAS and Enterprise Guide are still widely used, they are slowly but surely being pushed aside by web and cloud-based solutions such as SAS Studio and SAS Life Sciences Analytics Framework (LSAF). LSAF is a hosted solution that is designed to be much more than just a programming environment. It is an integrated data repository and compute environment offering features such as audit trail and versioning out-of-the-box.
Over the past year OCS Life Sciences has led the implementation of LSAF at a Belgian biotech company. This encompassed the (technical) configuration, design of the folder hierarchy and associated permission and security model, defining best practices for the use of the system, creation of jobs and process flows, and much more.
The LSAF environment in now in production and being utilised by data managers, programmers and statisticians. In this presentation we will share the most important learnings of our journey.

Martin Doughty - SAS Consultant - OCS Consulting
Daniel Christen - Principal Technical Account Manager - SAS Institue

"Moving your SCE to the cloud, focusing on Qualification and Validation"

For our customers in the life sciences industry, Computer Systems Validation is required by many Regulatory Agencies. These regulations require that all systems that govern any GxP process should be validated.
Validation applies to entire processes or analytical methods (and includes the equipment and software of computer system). It must follow industry best practices, regulatory standards, and defined processes. Qualification is a component of the overall validation process and provides documented evidence that the hardware and software comprising the system was installed (IQ) correctly and operates (OQ) as defined by the vendor for the customer's intended use.
This presentation will demonstrate for SAS Viya the Qualification approach in a Cloud Native environment and how it is designed to work with the customer's Standard Operating Procedures (SOPs) and Work Instructions (WIs) for Computer Systems Validation. Continuous Innovation / Continuous Delivery (CI/CD) will also be covered as part of this presentation.

Richard Paterson - Senior SAS Consultant - OCS Consulting

"SAS Coding Efficiency: Developing a Performance Mindset"

Coding efficiency is a major aspect when delivering software programs in any industry. We can think of efficiency being equal to reliability and speed of execution. In Life Science projects reliability of code is vitally important. With computers getting increasingly powerful is developing efficient code relevant? Is it perhaps more efficient to go with the first solution instead of creating efficient and performant code? If the code is run once the answer is Yes, however what happens when inefficient code is used in multiple projects and executed thousands of times? Or how will my code perform when processing increased amounts of data, will my code cope with processing large amounts of data from MRI scans, CT scans, wearables, gene sequencing?
This paper creates awareness for efficient programming within SAS. I will discuss efficient code design and strategies for utilizing the available computing resources. Finally I will discuss how to anchor these strategies by developing a performance mindset that places code efficiency at the forefront of software development.

Ruurd Bennink - Senior Statistical Programmer - OCS Life Sciences

"Differences in performance between the options KEEP and WHERE as compared  to the statements KEEP and IF in SAS data steps"

In general it is assumed that the KEEP option in a SAS data step is more efficient than using the KEEP statement. However this effect is more limited than expected and depends on the number of variables in the input dataset. An average dataset in life sciences (a million records and 50 variables) shows a mean performance gain of just a few percent points after keeping 2 variables.The same can be said after comparing the performance between the WHERE option vs. the IF statement. In this case the gain in performance depends on the number of selected observations. After selecting half of the records in a dataset of several million records and 50 variables the difference in performance is hardly detectable. After selecting 1% of the records the gain in performance is detectable, but again just a few percent points. This paper will look further into these results.

Stephan van Driel - SAS Consultant - OCS Consulting

"Using jobs and API macros to create a structured, controlled and personalised LSAF environment"

Once your company has decided to use LSAF, there are many things to consider, such as folder structures, permissions, groups, roles, role privileges, users accounts, user privileges and owners. Making the right choices made here lays the foundation for efficient working.
Just as important is the implementation of these steps in LSAF. While you can do everything manually this would be time consuming and prone to error, and it provides little insight into the environment’s current state.
Our recommended solution is to use the built-in LSAF API macros to automate all manual setup steps. Within OCS we have developed a set of standardised SAS programs (jobs in LSAF) and metadata containers that you can use to set up your own environment in a structured, controlled, personalised, faster and simple way.
In this presentation we will take you through the most important steps to automate the management of your LSAF environment.