Automation has the potential to utterly transform the efficacy of biological discovery; in cases where automation is well suited, this has been seen time and time again. However, there is a large category of biological experimentation that is substantially harder to automate— and for many scientists, automation still feels daunting and inaccessible. This talk will discuss how these fundamental problems can be overcome.
For experiments that vary substantially from one execution to another, automation is difficult to justify. This is because the investment required to establish the automated workflows is not paid back by the amount of usage those workflows will have before needing to be re-written. So to bring automation into these areas of high-variability science, we need a different way of programming automation. Current automation programming focuses on the movements of the robot (robot-oriented lab automation), but here we will introduce the concept of sample-oriented laboratory automation (SOLA). SOLA tracks the logic of how samples are treated in the course of an experiment, in a representation which reflects the science being done, rather than the movements a robot has to make. This more abstracted view of automated workflows enables much more rapid and flexible programming of automation, much like higher level programming languages enable software engineers to make much faster progress.
SOLA is essentially what underlies the Synthace platform, and here we will present the scientific benefits that it has brought in diverse use cases. In order to bring the power of a SOLA approach to these scientists, we have also deployed a still-higher level of abstraction, where commonly used SOLA methods are predefined. This takes advantage of the inherent agility of a SOLA approach to enable scientists to run very different variants of an experiment just by altering the parameters of a single SOLA workflow. This, combined with the ability of the Synthace platform to interface with a wide range of automation, gives a highly practical way for scientists to routinely use automation for their R&D.