Mining Microalgae To Reduce Carbon Emissions

A new machine learning-powered method is speeding up how scientists study microalgal cell factories and their carbon cycling functions.

AsianScientist (Nov. 24, 2021) – Powered by machine learning (ML) techniques, a novel method is accelerating the discovery of useful microalgae and their functions, including converting carbon dioxide into high-value molecules. The approach was reported in Analytical Chemistry by researchers from the Chinese Academy of Sciences’ Qingdao Institute of Bioenergy and Bioprocess Technology (QIBEBT).

Don’t let their size deceive you—microalgae may be small, simple and single-celled microorganisms, but they come in a huge variety of forms, each with diverse functions that could make a big dent in addressing societal issues like climate change.

Similar to how plants take up carbon dioxide to synthesize sugars, microalgae’s metabolic activities also naturally convert carbon dioxide into other compounds useful for producing food, fuels and drugs.

As communities explore initiatives to achieve carbon neutrality, such microalgal functions could potentially be harnessed to reduce carbon emissions across various chemical and manufacturing processes. But with the millions of microalgae inhabiting the planet, scientists have yet to figure out which species can most readily recycle carbon.

To pinpoint microalgal characteristics and metabolic activities, these organisms are typically grown in laboratory cell cultures, but this process is slow and tedious. A team at QIBEBT has developed a way to accelerate microalgal discovery, combining a chemical analysis technique called Raman microspectroscopy and a computing framework based on ML.

In Raman microspectroscopy, microalgae and their intracellular compounds scatter incoming high intensity light depending on their chemical structure and molecular interactions. The scattered Raman light waves are compiled into a spectra of signals that reveal each cell factory’s functions, producing two portraits—one from color-giving molecules and one from all other compounds.

While most Raman-based methods pick only one of the two images, the team combined the Raman portraits to uncover in-depth insights on miroalgae metabolism and built a database of Raman spectra from over 9,000 cells.

By using a ML algorithm to rapidly extract patterns from the data, the system identified species and their metabolic activities with a whopping 97 percent accuracy, tested against previously cultured microalgae.

Meanwhile, for uncultured microalgae, the researchers supplemented their strategy with genomic sequencing technology, analyzing the DNA of the species one cell at a time. With high-quality sequences, they could dissect the molecular world of each microalgal cell, leveraging the capabilities of ML to learn and improve as more data is fed into the system.

“This comprehensive approach for rapidly identifying and metabolically profiling single-cells, either cultured or uncultured, greatly accelerates the mining and screening of microalgal cell factories for carbon-neutral production,” said corresponding author Dr. Xu Jian from QIBEBT.

The article can be found at: Baladehi et al. (2021) Culture-Free Identification and Metabolic Profiling of Microalgal Single Cells via Ensemble Learning of Ramanomes.


Source: Chinese Academy of Sciences; Photo: Shutterstock.
Disclaimer: This article does not necessarily reflect the views of AsianScientist or its staff.

Asian Scientist Magazine is an award-winning science and technology magazine that highlights R&D news stories from Asia to a global audience. The magazine is published by Singapore-headquartered Wildtype Media Group.

Related Stories from Asian Scientist