As Lavi put it, one of the aims of her PhD is “to develop the first genomic resource to support future diagnostics and surveillance,” and that’s the gap her research fills: building the first reference genome for this pathogen. Getting there took most of her first year just building the raw materials to work with. Sugarcane stalks were sent by Dr Seona Casonato from Sugar Research Australia, and Lavi propagated them into seedling trays, only for the crop to be invaded by two-spotted spider mites before her fungus ever got a look-in. An anti-mite spray solved this problem. Lavi also travelled on a spore collection trip to Sugar Research Australia’s field station in Tully, Queensland, and then tested out controlled infection conditions from one relevant study she found, since brown rust has mostly been studied in the field. 

It worked, until her infection rate mysteriously slowed down. Detective work revealed that the very pesticide protecting her plants from mites, wettable sulfur, was also an effective fungicide, sneakily killing off the pathogen she was trying to grow. A pesticide swap and wiping down leaves before infection fixed the problem.

Then came the challenge of collecting the spores, since sugarcane plants are too large for the usual “bag and shake” method used for wheat rusts. The solution was assembling an in-house vacuum pump rig: tubing, a small spore collector, and an Eppendorf tube that let Lavi suction spores directly off infected leaves and store them for extraction. “This is a very DIY setup,” she said, “but I’m actually quite proud of it.”

From there, the project moved into genome assembly, combining Oxford Nanopore long-read sequencing with Hi-C data, a technique that maps which parts of the genome sit physically close together inside the nucleus. That distinction matters here because rust fungi are dikaryotic, with each cell carrying two separate nuclei, each with its own haploid genome, and Hi-C is what let Lavi cleanly separate the two.

Lavi filtered out contamination and mitochondrial DNA, to ensure she had DNA only from the brown rust pathogen. One piece of contamination got a laugh on its own: a small DNA sequence that turned out to be from Tetranychus urticae, the same two-spotted spider mite that had invaded her plants. “Probably because we used the vacuum pump, there were some leftover little mites, or eggs, that got suctioned in – it’s really hard to see them because they’re really tiny.”

The result, at least provisionally, is close to a complete chromosome-level assembly: 18 scaffolds per haploid nucleus, matching the number of chromosomes rust fungi are known to carry. The Hi-C data even revealed something interesting along the way – evidence that the chromosomes are sitting in a “Rabl configuration,” with centromeres clustered together in the nucleus, a structural feature that’s been studied in plants and animals but comparatively less in fungi.

Lavi’s next step is to aim for a full telomere-to-telomere assembly, and to layer in RNA-seq data to start understanding the biology behind the genome, not just its structure. This will create a high quality and biologically meaningful reference genome.

A reference genome gives researchers and diagnosticians a baseline to track emerging strains of brown rust against, supporting faster, more targeted surveillance as the pathogen evolves – and helping protect an industry where new resistance-breaking strains can be a devastating threat.

Thank you to Lavi, and to everyone in the Schwessinger Lab, the Training Centre, Sugar Research Australia and ANU who supported this work.