Biomedical research workflows currently suffer from a lack of tools enabling the collection of the experimental context. The multivariate nature of the experimental context makes it intractable to continuously record the evolution of all the variables over the length of the experiment. As a result, many variables are never recorded which turns the debugging of a failed protocol into a guessing game. This not only makes the discovery of significant experiment variables difficult, but also constitutes the main barrier in training new individuals on these workflows.
The lack of contextual data is a key factor in the current inability to apply a big data approach to the biomedical experimental workflow. Currently, the only way to create sufficient data to enable such an approach requires the use of expensive and inflexible automation platforms. These platforms require significant adaptation of the workflows that only make it worthwhile if large numbers of samples are used or if they are made of multiple repeated operations.
We propose to create a framework enabling the rapid prototyping of biomedical workflows through the seamless collection the experimental context and creation of personalized tools for experimentation. Using rapid prototyping fabrication tools, we aim to develop a suite of open and lowcost hardware to sense the experiment’s contextual data (bench and sample temperature, CO2 concentration, etc) as well as perform common laboratory operations.
We believe this framework and tools have the potential to transform biomedical research and teaching workflows by enabling big data insights into experimental data.