Artificial Intelligence for Genomic Selection of Sugarcane in Fields Developed in Brazil

Scientists suggested using Artificial Intelligence for genomic selection to better predict the efficient performance of sugarcane in fields.

Brazil is one of the main producers of sugarcane worldwide. The sugarcane industry has been a source of livelihood for many farmers.

According to the Britannica website, sugarcane, also known as Saccharum officinarum, grows in subtropical and tropical areas worldwide. Sugarcane products have many uses, including to produce of ethyl alcohol or ethanol, cane fibers, straw, rum, and others used for biofuel.

Furthermore, the Brittanica website explained that sugarcane requires a condition soil to propagate or prosper. Based on records, sugarcane crop requires well-drained soil, a mixture of sand and clay particles, which could also grow in alluvial soils nearby rivers and volcanic soils. It is also important that sugarcane has 80 to 90 inches of water to attain a good harvest.

The research was published in Scientific Reports and is available on the Phy.org website.

Artificial intelligence

As a result, new research emerged to help farmers predict the performance of sugarcane in agricultural fields. For this time, scientists in Brazil developed the application of artificial intelligence to create efficient models for the sugarcane's genomic selection, noting to protect the sugarcane's performance output.

Furthermore, the use of Artificial Intelligence (AI) has been well-known in many fields. AI is a branch of machine learning and computer science providing useful statistics for optimization, production, and intervention. AI could provide insights into the effect of pests, insects, bacteria, fungi, soil nutrients, droughts, and climate change on the sugarcane harvest.

According to the report, the application of AI would help predict sugarcane performance based on their DNA. Amazingly, the results showed that the methodology using IA showed a 50% improvement in the predictive results for sugarcane. The report also noted that the efficient genomic selection method was the first time in terms of polyploid plants.

Efficient to predict

The study's first author and computer scientist, Alexandre Hild Aono, explained that crossing had been used for traditional breeding. He explained that the process included crossing sugar-producing sugarcane to resistant sugarcane. Aono is also a researcher from the State University of Campinas's Center - Molecular Biology and Genetic Engineering.

The researchers noted that the shared method could predict the plants' performance even before they started to grow, noting that the assessment was based on genetic material. Interestingly, the technique can save years of evaluation.

However, he added that the assessment would take time and be expensive, incurring a high cost for field owners and farmers. Meanwhile, Aono noted that the genome complexities of the sugarcane and forage grass are one hurdle to breeding polyploid plants.

Moreover, Anete Pereira de Souza said that breeders identify interesting plants and clones, but it requires more time and cost. However, Souza noted that intensive use of data and efficient statistical tools (Plant breeding 4.0) would help overcome the hurdles. Souza is also a professor of plant genetics from UNICAMP's Institute of Biology.

For more similar, don't forget to follow Nature World News.