RNA Sequencing data that pointed the wrong way: how a researcher used EMET to understand contradictory data
Dr. Emanuela Pania is a postdoctoral researcher at The Hospital for Sick Children in Toronto, where she studies gene environment interactions in neuromuscular disease. Her work runs wide by necessity, from molecular analysis and RNA sequencing to behavioural testing and controlled immunological and dietary exposures. The questions she asks sit at the boundary between a genome and everything acting on it, which means the evidence she needs is rarely in one place.
She came to EMET having already worked with other AI tools, so she had a basis for comparison. What stood out first was the quality of what came back.
"It thinks better than a lot of other AI platforms in the sense of the literature, and from genes to pathway analysis."
A pathway enrichment result that pointed the wrong way
One of her models had produced something she could not immediately explain. A pathway she expected to be strongly upregulated came back with all of her target genes downregulated. She pasted the data into EMET and asked for pathway enrichment analysis. EMET returned the pathways she expected, corroborating her own reading. She then told it the direction of regulation was inverted, and it surfaced published work reporting the same downregulation.
"It brought my analysis to a new level, without me having to spend hours searching for that one paper that reports a downregulation of a pathway that was expected to be up - for example."
That is the part a keyword search does not do. The relevant paper exists, but it is old, it is not highly cited, and it contradicts the expected direction, so it does not surface. On what the evidence gathering was worth to her:
"It's essentially you doing weeks of your own research into the topic in a session."
The pathway maps she did not ask for
EMET chose the format without being asked, and drew the biology rather than describing it.
"It generates these pathway maps for you without you asking for them. I'm a very visual learner. It's one thing to have paragraphs and paragraphs of words that you just have to read in detail. It's another thing to have a visual representation right there."
Not another enrichment tool
The enrichment result is the starting point, not the answer. What she valued was EMET reasoning through what the result meant for her biology.
"It's not just another pathway enrichment tool. It explains what those pathways mean, which for someone who isn't used to seeing those pathways is very helpful. It is like having a scientist beside you saying, hey, this is your hit and this is what it means."
Where the acceleration shows up
Asked whether any of this had accelerated her projects, she details the potential of having EMET at her fingertips:
"It's essentially you doing weeks of your own research into the topic in a session. So yes, I do think it can enable faster development of experiments that address the question you're trying to answer."
"I definitely think the connections are there at your fingertips."
That is where the time actually goes in a lab like hers. Not in the sequencing run, but in the weeks between a confusing result and a well designed follow up, spent reading toward a paper you might never find. Compress those weeks into a session and the next experiment is sharper before anyone touches a pipette.
The pathway was still downregulated when EMET was done with it. What changed was how quickly she got from an anomaly she could not explain to a testable idea about why.
