Design, deliver, and run enterprise blockchain workloads quickly and easily.
All servers and systems
The world is becoming a hungrier place. Its population is expected to grow to 9 billion by 2050, according to the United Nations, and feeding everyone will require substantial increases in global food production. But many factors, including climate change, pests, and disease, make that increasingly difficult, given existing agricultural techniques and knowledge. In fact, a study at the University of Minnesota notes that the world currently isn’t on track to meet crop production goals.
But scientists say there is hope to help meet the challenges: a new agricultural revolution that uses genomics and bioinformatics, along with new genome-editing techniques. It’s a revolution being played out in universities and research centers around the globe, and the weapons are powerful supercomputers and advanced genetics techniques.
The revolution is being fed by the explosion of data about plant and animal genomes, and the analysis of it all via bioinformatics, the science of analyzing complex biologic data. Andrew Severin, a research scientist and manager of the Genome Informatics Facility (GIF) at Iowa State University, points to the mapping of the human genome by the Human Genome Project in 2001 as ushering in the era of bioinformatics.
“The amount of data created since then has exploded,” Severin says. “It quadruples roughly every year. The cost of sequencing genomes continues to decrease, and the amount of data generated continues to increase.”
All that data can yield genetic treasure, not just to help feed the world, but also for human health and beyond. However, Severin says, “A lot of biologists and scientists haven’t been trained to handle that quantity of data. So the new field of bioinformatics had to be created to process and analyze it in a meaningful way.”
Bioinformatics requires high-power computing, as does sequencing the genomes of plants and animals and identifying the genetic roles played by individual genes and gene variants. At GIF, Severin says, the computing infrastructure to do it all uses 100,000 CPU hours each month, equivalent to running 60 dual-core laptops simultaneously every hour, every day of the month.
All that power means more genomes can be sequenced more quickly at far less cost. Severin says that in 2001, it cost $5,000 to sequence a section of genome segment called a megabase. Today, it costs pennies.
To understand how these new techniques can help boost food production around the world, it’s worth taking a look back at our agricultural past. Many of the crops that provide our food today were bred from wild species that were common thousands of years ago. So their genomes are different today than they were back then.
The cost of sequencing genomes continues to decrease, and the amount of data generated continues to increase.
Arun Seetharam, a scientist at GIF, explains, “The genomes of crops have undergone many changes over time to make them human-friendly. Take corn, for example. The cobs were originally very small, but we bred them to make the cobs larger to provide us with more nutrition.”
However, that breeding has caused a variety of problems as well. There is less genetic variety in modern crops than older ones, because fewer variants are planted over wide areas, making them susceptible to pests, diseases, and a changing climate, Seetharam says. “Diversity has been reduced, and so if the environment changes even a little bit, it can cause big problems with crops. They can’t adapt to the changes like their wild relatives do, because they’ve lost the genes they need for survival in a changing world.”
One answer to the problem, he says, is to analyze the genomes of the original wild varieties “and see what genes are useful now, so we can transfer some of the traits from their wild relatives that can survive insects and pathogens.”
It’s not only plants that can benefit from gene sequencing and bioinformatics. Severin’s recent research has focused largely on aquaculture. Fish haven’t been bred to the same extent as crops such as corn have, and so he doesn’t look for genes in ancient, wild varieties. Instead, he looks for things such as genetic variants that can lead to problems that might make fish less suitable for eating and then breeds that gene out of them.
The key to all this depends on creating a map of the entire genome of a plant or animal, identifying how the genes work in it, and then altering the genome in beneficial ways. Because genomes of plants and animals can be mapped so much more quickly and at less expense than previously, the genomes of many more individuals can mapped. That gives scientists more data in which to find useful gene variants.
However, finding the functions of genes and variants is one thing, while using that information to help feed a hungry world is another. Scientists have to make sure the useful gene variant makes its way into the target plant or animal.
Traditional breeding is one way. In that case, when researchers find two individuals with the same beneficial gene variant, they breed them, so that their offspring will have the variant. In this way, they create an entire line with that useful variant. This is the technique Severin uses.
The other way is a technique known as CRISPR. In essence, it lets scientists edit the genome of a plant or animal by directly modifying their genes without any breeding. The genome with the beneficial gene of the plant or animal will be passed to its offspring.
Severin notes that using this technique, one could identify a gene variant that increases the protein content of fish and modify that gene, and then the fish and its offspring would then produce more protein—and so feed more people.
Some of the most useful work of this kind is being done by the Earlham Institute in the U.K. It works on a wide variety of projects aimed at improving human, animal, and plant health, and improving global food security. The institute has developed a reputation as an expert in next-generation genomics and bioinformatics, and it collaborates with other institutions around the world.
One of Earlham Institute's most important projects is piecing together the genomes of a number of varieties of wheat. The institute targeted wheat because it is a staple food for more than 2 billion people around the world, and it is grown on more land than any other commercial crop. If wheat can be hardened against pests, drought, and heat, and grown at less cost more reliably, it would go a long way toward helping feed a hungry world. The institute is particularly interested in producing hardier and more diverse varieties of wheat that offer higher yields and can better adapt to environmental changes. It wanted not only to sequence the DNA of many wheat varieties, but also identify important genes and gene combinations and model how they work.
The institute targeted wheat for another reason: It is an extremely complex genome, larger than any common commercial crop and five times the size of the human genome, making it difficult and time-consuming to map.
The work done at the Earlham Institute makes clear how vital an increase in computing power is in making breakthroughs in genomics. Doing the work necessary for assembling a wheat genome requires between 6 and 12 terabytes of RAM. It took the institute almost a month to complete. To speed up the work, it installed the SGI UV 300 supercomputer platform — the largest one in use for the life sciences, and one of the largest deployments of it worldwide.
The institute used two SGI UV supercomputers to map wheat genomes, and the results were dramatic: an 80 percent increase in code running speed, and a significant decrease in the time it takes to assemble a wheat genome—from four weeks to only days. That means scientists can more quickly identify genes that can help develop better wheat varieties to improve yield and create hardier wheat.
Chris Burt, a molecular breeder at RAGT Seeds in Cambridgeshire, England, told the Eastern Daily Press that mapping the wheat genome “enables us to be a lot more accurate and smarter in the way we can identify genes, and combine the beneficial traits that farmers and consumers are looking for. We can be more precise in what we are doing, so farmers can get the bread-making characteristics, combined with high yields and disease resistance, so they can produce more reliable crops.”
The Earlham Institute uses the techniques for more than just wheat. Among other crops, it has helped sequence the genome of the white Guinea yam, a food staple in Africa that millions of people depend on. The institute’s Benjamen White says, “Having a genomic resource for white Guinea yam will be invaluable in breeding a better yam, one that will improve food security in West and Central Africa, and the livelihood of smallholder farmers there."
Looking to the future, GIFs Severin believes that high-power computing techniques like those used at GIF and the Earlham Institute will help feed a growing world. “We need to do better in the coming years to feed the world and making sure people don’t go hungry,” he says. “Bioinformatics and genomics is an important way to do that. In the future, sustainable aquaculture, sustainable agriculture, and sustainable livestock is going to become increasingly important because we have shrinking land areas on which we can perform agriculture, aquaculture, and maintenance of livestock. With the new techniques, I’m optimistic we’ll be able to do that.”
This article/content was written by the individual writer identified and does not necessarily reflect the view of Hewlett Packard Enterprise Company.