Kiwi scientists have designed a new way to unravel the genomic origins of "tricky" Covid-19 cases, which could shed fresh light on Auckland's August outbreak.
Genome sequencing creates a "genetic fingerprint" of a virus that has infected a person, and can help public health officials untangle different cases involved in an outbreak through their genetic sequences.
In New Zealand's first wave of Covid-19, scientists sequenced the genomes of 649 separate cases to reveal nearly 300 different introductions from different parts of the world.
Sequencing proved just as crucial in the August outbreak, helping pick apart Auckland community cases – effectively informing the response to the cluster in real time.
Yet not all of those could be sequenced. Of 175 samples sent to scientists, 145 could be fully sequenced, while 20 yielded only partial results.
And while scientists were still able to extract enough information from 15 of those to link them with the B.1.1.1 lineage that dominated the outbreak, others couldn't be reliably sequenced.
Matt Storey, a bioinformatics scientist at ESR, described that shortfall as a "signal to noise ratio issue".
What did that mean?
For genome sequencing to be successful, he explained, there needed to be some intact regions of the viral RNA genome present in the sample.
RNA was a fragile molecule that could degrade quickly when damaged by light, heat and chemicals - and such could be expected with samples, even when handled and stored in ideal conditions.
The sequencing strategy ESR scientists have adopted required RNA fragments to be at least 500 nucleotide - or sequence "letters" - in length for a sequence to be generated.
But when the concentration of the RNA in a sample was very low, the probability of finding a fragment long enough for a given region of the genome began to drop below the level of detection.
"Furthermore, when the viral RNA is extracted, all of the RNA from the human cells and any other stowaways in the clinical sample is also extracted along with it, this means there is a lot of background RNA to contend with."
In samples with low amounts of the virus RNA, the "background" RNA could interfere too much with the chemical processes needed to prepare the sample for sequencing.
Like the way Covid-19 was detected and confirmed from samples taken from swab tests, the sequencing process relied on polymerase chain reaction, or PCR - but with some differences.
When sequencing, scientists targeted the entire genome in many parts, not just some small selected part as the diagnostic test did, which made the process much more complex and sensitive.
This meant that in "weak" positive samples, some parts of the genome could simply be degraded beyond the level that sequencing could do its full job.
"If there is little virus, we might fail to extract a genome completely, but we also might be able to extract only a partial genome," Storey said.
"This can still provide clues for us on lineage or type of that virus but not the full picture."
For these partial genomes, the usefulness of the data depended on which parts they could retrieve.
"It's kind of like only getting a few pieces of the jigsaw puzzle and trying to figure out what the complete picture is," he said.
"You might get lucky with the right parts, like an eye or an ear in a photo, and you might not, like a piece of the white border."
Now, Storey and his colleagues think they've found a new method enabling them to solve these incomplete puzzles.
That's in a more sensitive assay that could work on lower concentrations and shorter pieces of RNA, when the team knew what mutations to look for.
"It will allow us to more thoroughly investigate certain cases where we were previously unable to generate a genome."
Their assay would be targeted at samples with low viral loads, or high cycle threshold or "CT" values - as many cases from managed isolation had.
"This new assay will give us a greater likelihood that samples with high CT values will produce a workable genome.
"This produces more information for the response, like epidemiologists when working on tricky case investigations.
"We don't yet know where the current Auckland August cluster originated, but we are continuously evaluating the data," he said.
"Every time a new piece of evidence pops up, we incorporate it into the analysis and investigate it, so this may shed new light."
It comes after a top-level review described New Zealand's genome sequencing effort as "world-leading" – but has still found room for improvement.
It found that missing genomic data from positive samples would continue to hamper efforts to pinpoint the sources of any outbreaks that might escape border quarantine.
It also recommended that samples which failed to yield full viral genomes should be analysed using shorter fragments of the genomes, ensuring that a viral strain could still be determined from even small sections of the virus.