Machine learning could enable the bioengineering of the most abundant enzyme on the planet
Newswise – A Newcastle University study has for the first time shown that machine learning can predict the biological properties of the most abundant enzyme on Earth – Rubisco.
Rubisco (Ribulose-1,5-bisphosphate carboxylase/oxygenase) is responsible for providing carbon for almost all life on Earth. Rubisco works by converting atmospheric CO2 from the Earth’s atmosphere into organic carbonaceous matter, which is essential for sustaining most life on Earth.
For some time, natural variations have been observed among Rubisco proteins in land plants and modeling studies have shown that transplanting Rubisco proteins with certain functional properties can increase the amount of atmospheric CO2 that crop plants can absorb and store. .
The study’s lead author, Wasim Iqbal, a PhD researcher at Newcastle University’s School of Natural and Environmental Sciences, part of Dr Maxim Kapralov’s group, has developed a machine learning tool that can predict the performance properties of many land plant Rubisco proteins with surprisingly good accuracy. The hope is that this tool will enable the search for a “supercharged” Rubisco protein that can be bio-engineered into major crops such as wheat.
Posted in the Journal of Experimental Botany, the study presents a useful tool for screening and predicting plant Rubisco kinetics for engineering efforts as well as fundamental studies of Rubisco evolution and adaptation. Screening for the natural diversity of Rubisco kinetics is the main strategy used to find better Rubiscos for crop engineering efforts.
Wasim said: “Our study will have huge implications for climate models and bioengineering crops.
“This study provides plant biologists with a screening tool to highlight Rubisco species with better kinetics for crop engineering efforts.
“The machine learning tool can be used to improve the accuracy of global estimates of photosynthesis. The Rubisco performance properties predicted by our model are consistent with Earth System Models (ESMs) used by climatologists. Currently, ESMs use a single set of Rubisco properties from the same species (or sometimes a handful) to estimate ecosystem-scale photosynthesis. Our machine learning tool could provide predictions for most land plants, improving the accuracy of ESMs. »
The next steps in this work include isolating the best Rubisco proteins identified from laboratory predictions and attempting to bioengineer a plant species with a foreign Rubisco protein.
Iqbal, W., Lisitsa, A. and Kapralov, M. (2022). Predict Rubisco kinetics of plants from RbcL sequence data using machine learning. Journal of Experimental Botany. doi: 10.1093/jxb/erac368