news

Researchers accelerate organoid development and selection using algorithm

The semi-automated process enabled researchers to make retinal organoid production and selection nearly four times faster.

robotic hand pipetting from petri dish of organoids next to a clock - idea of automation speeding up organoid development [Credit: Daria Sokol/MIPT Press Office].

[Credit: Daria Sokol/MIPT Press Office].

Scientists have made the process of producing quality stem cell-derived tissues for research nearly four times faster using an algorithm. While their paper describes its use in accelerating retinal organoid development, the researchers suggest their algorithm could be adapted for other in vitro tissue cultures as well.

The retina is the portion of the eye that senses light and pre-processes it ahead of signals being sent to the brain. It is a tissue consisting of several layers of neurons forming a chain. Due to their limited ability to regenerate, loss of these neurons results in permanent blindness. Approximately 15 million Americans have retinal degenerative diseases, but the number is growing across the world, primarily because of population aging.

To develop therapeutic strategies for retinal degeneration, whether they be pharmacological neuroprotection, gene therapy or cell replacement, massive amounts of retinal tissue are required for research.

While retinal tissue can be grown successfully in vitro, researchers say the random nature of the initial neuron growth stimulation and the inconvenience of the time taken for artificial retinas to develop correctly (30 days for murine retinas and up to one year for human organoids) severely limit the throughput of traditional methods.

To address these problems, researchers from Rusia’s Moscow Institute of Physics and Technology (MIPT) and Harvard University, US, semi-automated the process of tissue differentiation, using an algorithm to increase the number of cells produced and their quality.

To compare the quality of the robot- and human-grown cells they produced several thousand retinal tissue samples for automatic processing and the same amount for manual handling. The researchers scanned the wells housing the developing tissue samples and used a Python script they wrote to analyse the images. According to the team the programme determines the areas in the photos where the glow of the fluorescent protein is the strongest. Since this protein is only produced in developing retinal cells, the points with high fluorescence intensity are the areas of the sample where retinal cells are growing. This allows the software to determine the amount of developing retina in each organoid.

The researchers reported that the automation algorithm was able to optimise cell production without any changes in differentiation pattern or adverse effect on tissue quality. The new approach reduced the time researchers needed for cell processing from two hours to just 34 minutes for 960 organoids over the course of 25 days.

Evgenii Kegeles of the Genome Engineering Lab at MIPT said: “We… developed a tool for automatic retina identification and organoid classification, which we showed in action, optimising cell specialisation conditions and monitoring tissue quality.

“One of our goals in this research has been to scale up cell differentiation to enable a high-throughput tissue production for drug tests and cell transplantation experiments. Automated sample handling makes it possible to reduce the effort on the part of the personnel and produce several times more cells in the same period of time. With some slight modifications, the algorithm would be applicable to other organoids, not just the retina.”

Petr Baranov from the Schepens Eye Research Institute of Massachusetts Eye and Ear, US, commented: “Numbers matter: the automation empowered us to produce trillions of retinal neurons for transplantation and we are excited to see the translation of our approach into routine cell manufacture.”

The study was published in Translational Vision Science & Technology.