AI reveals hidden traits about our planet's flora to help save species

In a world-first, scientists from UNSW and Botanic Gardens of Sydney, have trained AI to unlock data from millions of plant specimens kept in herbaria around the world, to study and combat the impacts of climate change on flora.

"Herbarium collections are amazing time capsules of plant specimens," says lead author on the study, Associate Professor Will Cornwell. "Each year over 8000 specimens are added to the National Herbarium of New South Wales alone, so it's not possible to go through things manually anymore."

Using a new machine learning algorithm to process over 3000 leaf samples, the team discovered that contrary to frequently observed interspecies patterns, leaf size doesn't increase in warmer climates within a single species.

Published in the American Journal of Botany, this research not only reveals that factors other than climate have a strong effect on leaf size within a plant species, but demonstrates how AI can be used to transform static specimen collections and to quickly and effectively document climate change effects.

Herbarium collections move to the digital world

Herbaria are scientific libraries of plant specimens that have existed since at least the 16th century.

"Historically, a valuable scientific effort was to go out, collect plants, and then keep them in a herbarium. Every record has a time and a place and a collector and a putative species ID," says A/Prof. Cornwell, a researcher at the School of BEES and a member of UNSW Data Science Hub.

A couple of years ago, to help facilitate scientific collaboration, there was a movement to transfer these collections online.

"The herbarium collections were locked in small boxes in particular places, but the world is very digital now. So to get the information about all of the incredible specimens to the scientists who are now scattered across the world, there was an effort to scan the specimens to produce high resolution digital copies of them."

The largest herbarium imaging project was undertaken at the Botanic Gardens of Sydney when over 1 million plant specimens at the National Herbarium of New South Wales were transformed into high-resolution digital images.

"The digitisation project took over two years and shortly after completion, one of the researchers -- Dr Jason Bragg -- contacted me from the Botanic Gardens of Sydney. He wanted to see how we could incorporate machine learning with some of these high-resolution digital images of the Herbarium specimens."

"I was excited to work with A/Prof. Cornwell in developing models to detect leaves in the plant images, and to then use those big datasets to study relationships between leaf size and climate," says Dr Bragg.

"Computer vision" measures leaf sizes

Together with Dr Bragg at the Botanic Gardens of Sydney and UNSW Honours student Brendan Wilde, A/Prof. Cornwell created an algorithm that could be automated to detect and measure the size of leaves of scanned herbarium samples for two plant genera -- Syzygium (generally known as lillipillies, brush cherries or satinas) and Ficus (a genus of about 850 species of woody trees, shrubs and vines).

"This is a type of AI is called a convolutional neural network, also known as Computer Vision," says A/Prof. Cornwell. The process essentially teaches the AI to see and identify the components of a plant in the same way a human would.

"We had to build a training data set to teach the computer, this is a leaf, this is a stem, this is a flower," says A/Prof. Cornwell. "So we basically taught the computer to locate the leaves and then measure the size of them.

"Measuring the size of leaves is not novel, because lots of people have done this. But the speed with which these specimens can be processed and their individual characteristics can be logged is a new development."

A break in frequently observed patterns

A general rule of thumb in the botanical world is that in wetter climates, like tropical rainforests, the leaves of plants are bigger compared to drier climates, such as deserts.

"And that's a very consistent pattern that we see in leaves between species all across the globe," says A/Prof. Cornwell. "The first test we did was to see if we could reconstruct that relationship from the machine learned data, which we could. But the second question was, because we now have so much more data than we had before, do we see the same thing within species?"

The machine learning algorithm was developed, validated, and applied to analyse the relationship between leaf size and climate within and among species for Syzygium and Ficus plants.

The results from this test were surprising -- the team discovered that while this pattern can be seen between different plant species, the same correlation isn't seen within a single species across the globe, likely because a different process, known as gene flow, is operating within species. That process weakens plant adaptation on a local scale and could be preventing the leaf size-climate relationship from developing within species.

Using AI to predict future climate change responses

The machine learning approach used here to detect and measure leaves, though not pixel perfect, provided levels of accuracy suitable for examining links between leaf traits and climate.

"But because the world is changing quite fast, and there is so much data, these kinds of machine learning methods can be used to effectively document climate change effects," says A/Prof. Cornwell.

Read more at Science Daily

Researchers find genetic links between traits are often overstated

Many estimates of how strongly traits and diseases share genetic signals may be inflated, according to a new UCLA-led study that indicates current methods for assessing genetic relationships between traits fail to account for mating patterns.

Through the use of powerful genome sequencing technology, scientists in recent years have sought to understand the genetic associations between traits and disease risk, hoping that discoveries of shared genetics could point to clues for tackling diseases. However, UCLA researchers said their new study, published Nov. 17 in Science, provides caution against relying too heavily on genetic correlation estimates. They say that such estimates are confounded by non-biological factors more than has been previously appreciated.

Genetic correlation estimates typically assume that mating is random. But in the real world, partners tend to pair up because of many shared interests and social structures. As a result, some genetic correlations in previous work that have been attributed to shared biology may instead represent incorrect statistical assumptions. For example, previous estimates of genetic overlap between body mass index (BMI) and educational attainment are likely to reflect this type of population structure, induced by "cross-trait assortative mating," or how individuals of one trait tend to partner with individuals of another trait.

The study authors said genetic correlation estimates deserve more scrutiny, since these estimates been used to predict disease risk, glean for clues for potential therapies, inform diagnostic practices, and shape arguments about human behavior and societal issues. The authors said some in the scientific community have placed too much emphasis on genetic correlation estimates based on the idea that studying genes, because they are unalterable, can overcome confounding factors.

"If you just look at two traits that are elevated in a group of people, you can't conclude that they're there for the same reason," said lead author Richard Border, a postdoctoral researcher in statistical genetics at UCLA. "But there's been a kind of assumption that if you can track this back to genes, then you would have the causal story."

Based on their analysis of two large databases of spousal traits, researchers found that cross-trait assortative mating is strongly associated with genetic correlation estimates and plausibly accounts for a "substantial" portion of genetic correlation estimates.

"Cross-trait assortative mating has affected all of our genomes and caused interesting correlations between DNA you inherit from your mother and DNA you inherit from your father across the whole genome," said study co-author Noah Zaitlen, a professor of computational medicine and neurology at UCLA Health.

The researchers also examined genetic correlation estimates of psychiatric disorders, which have sparked debate in the psychiatric community because they appear to show genetic relationships among disorders that seemingly have little similarity, such as attention-deficit hyperactivity disorder and schizophrenia. The researchers found that genetic correlations for a number of unrelated traits could be plausibly attributed to cross-trait assortative mating and imperfect diagnostic practices. On the other hand, their analysis found stronger links for some pairs of traits, like anxiety disorders and major depression, suggesting that there truly is at least some shared biology.

"But even when there is a real signal there, we're still suggesting that we're overestimating the extent of that sharing," Border said.

Read more at Science Daily

Plant breeding: Using 'invisible' chromosomes to pass on packages of positive traits

The ideal crop plant is tasty and high-yielding while also being resistant to diseases and pests. But if the relevant genes are far apart on a chromosome, some of these positive traits can be lost during breeding. To ensure that positive traits can be passed on together, researchers at Karlsruhe Institute of Technology (KIT) have used CRISPR/Cas molecular scissors to invert and thus genetically deactivate nine-tenths of a chromosome. The traits coded for on this part of the chromosome become "invisible" for genetic exchange and can thus be passed on unchanged. The researchers have reported on their findings in Nature Plants.

Targeted editing, insertion or suppression of genes in plants is possible with CRISPR/Cas molecular scissors. (CRISPR stands for Clustered Regularly Interspaced Short Palindromic Repeats.) This method can be used to make plants more resistant to pests, diseases or environmental influences. "In recent years, we were able for the first time to use CRISPR/Cas not only to edit genes but also to change the structure of chromosomes," says Professor Holger Puchta, who for 30 years has been researching applications for gene scissors with his team at KIT's Botanical Institute. "Genes are linearly arranged along chromosomes. By changing their sequence, we were able to show how desired traits in plants can be separated from undesired ones."

Now the researchers have been able to prevent the genetic exchange that is normally part of the hereditary process but can break the links between traits. "We can shut down a chromosome almost completely, making it seem invisible, so that all traits on that chromosome can be passed on in a package," says Puchta. Until now, if a plant's traits were to be passed on together, the genes for those traits needed to be close to each other on the same chromosome. If such genes are spread farther apart on a chromosome, they are usually separated during inheritance, so a beneficial trait can be lost during the breeding process.

Learning from Nature: Chromosome Engineering Prevents Genetic Exchange

In their research, the scientists followed nature's example. "These reversals, or inversions -- a kind of genetic invisibility -- also occur frequently on a smaller scale in wild and cultivated plants. We've learned from nature and have applied and extended our knowledge about the natural process," says Puchta.

In collaboration with Professor Andreas Houben from the Leibniz Institute of Plant Genetics and Crop Plant Research (IPK), Puchta and his team inverted nine-tenths of a chromosome in the model organism Arabidopsis thaliana (thale cress). Only at the ends of the chromosome did the genes retain their original sequence. "With these fragments, the chromosome can be passed on to the next generation just like the other chromosomes and is not completely lost," says Puchta.

Read more at Science Daily