EFRI BEGIN OI Teaching non-brain organoids how to think: PRogrammable OrGanoid intElligence using neuronal Networks implemented by gene Circuits (PROGENIC)

The brain is the most prominent biological computer. As a result, efforts to develop biocomputing systems have focused on using neurons as their basic building block. That approach misses the fact that every organ in the body continuously senses and responds to a variety of inputs.

These inputs can be a combination of electrical, chemical, mechanical or optical. These organs make decisions (i.e. compute responses) based on those inputs. With that in mind, the project team will attempt to develop intelligent organ-like masses of cells (organoids) based on liver cells.

The hope is that understanding how to develop functional intelligence in this system will make it possible to do so using any type of cells. The notion of creating intelligent biological systems gives rise to many ethical questions. Questions related to whether such systems could evolve a form of consciousness will be addressed.

Focused participation in an annual Science Olympiad for middle school students will be a primary effort to broaden the participation of students in hands-on science and engineering. It is expected that this will lead to greater involvement of those students in STEM careers at all levels.

PROGENIC enables an intelligent liver organoid to learn and adapt to environmental conditions like viral infections and toxins. Engineered cells in the organoid are equipped with genetically encoded artificial neuronal networks that operate using DNA, RNA, and proteins. These cells are designed to learn new functions by modifying the genetically-encoded weights within their artificial neural networks.

This is implemented using on-demand DNA modifications to genetic circuits in the chromosomes of living cells and is induced by communication from designer guide cells carried to various locations in the 3D organoids by magnetically controlled microrobots. Assessment of these new neuronal networks is performed by exposing the organoids to variable environmental conditions and perturbations, including infections and toxins.

After these perturbations/insults, intelligence is evaluated by quantifying organoid function (e.g. albumin and urea secretion). The project is structured into three aims. Aim 1 engineers liver cells to harbor multi-node artificial neuronal networks within each cell, using synthetic gene circuits that operate with DNA, RNA, and proteins.

Aim 2 utilizes microscale robots to facilitate communication with cells in organoids that change neural circuit weights and enable learning. Aim 3 creates an intelligent liver organoid that modifies organoid structure and function and evaluates the performance of the biocomputing system. In addition, neuroethical issues, such as the possibility of consciousness and sentience arising in the de novo neuronal network, are also explored.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.