Papers
Each has a plain summary under the title. A preprint is a paper posted publicly before peer review, so the reasoning can be checked by anyone rather than only by two anonymous reviewers.
Each manuscript is readable in full here, until it is posted to a preprint server and receives a permanent identifier.
The method I had been relying on to predict which variants break this channel only works when the cause is a buried charge. Given four variants already known to break it, the method caught one. This is the result that closed the route I most wanted to work.
Closed the folding-corrector route.
Counting the places a base editor might act by mistake, by asking whether it could actually edit there rather than whether the sequence merely looks similar, removes 85.1 percent of the apparent risk. It also shows that the naive method is biased toward recommending the worse of two editors.
The method is constructive. The finding that no guide at this site is clean in an absolute sense is not.
Underpins the base-editing route.
Laboratories deposit real measurements of what variants do into the public database. None of the eleven determinable interpretation tools reads that field, and most of them already have the full record on disk. The evidence is public and effectively invisible.
The database census, and the one result that is not about this variant.
Every design I tested for editing the message rather than the gene either changed neighbouring letters it should not have, or bound in too many other places. Separately, no molecule of this class has been shown to reach a human heart muscle cell at all.
Sets the ceiling on the RNA-editing route.
My variant is labelled Conflicting in the public database, which reads as genuine disagreement between laboratories. It is not. It rests on a single stale submission. Across every available assay it behaves the same as its neighbour, which is classified pathogenic.
The reclassification argument.
A therapy that boosts output from the healthy copy needs a reserve of wasted message to redirect. Across 827 heart samples that reserve is 0.0045 percent, about 300 times smaller than in the brain where the approach already works in children. This closes the route.
Closed the upregulation route.
The standard assay cannot tell apart two different ways a broken copy interferes with a healthy one, because the two explanations are algebraically degenerate: they predict the same number. Measuring harder with the same method cannot help. This paper names the experiment that would settle it.
Names the experiment that would settle the open question.
Two candidate explanations for why this position matters were tested and eliminated. What survives is that this residue anchors a buried acidic pocket unchanged across species. The paper also corrects several of my own earlier claims about the geometry.
Eliminates two explanations for the mechanism.
Of 131 variants of uncertain significance in this region, 17 score above a measured noise floor and are worth testing. The other 114 are uninformative, which is not the same as benign. The blind spots of the predictor are named in the title deliberately.
A usable shortlist, built on top of the measured blind spot that makes a negative score meaningless.
Nominates seventeen other variants for testing.
Models that measure distance between the centres of residues miss about 9 percent of real atomic contacts, and disproportionately the long side-chain and salt-bridge ones. The contact at my own position is 3.79 angstroms atom to atom but 9.22 centre to centre, so a centre-distance model cannot see it at all.
A methods correction to a published penetrance model.
Seven of these report that something did not work. One closed the route this project had spent months on and most wanted to succeed. Another found that a prediction method I had been relying on identified one of the four cases it should have caught, which means a reassuring score from it carries no information at all in this part of the gene.
A record that reports only what worked is not a record, it is a selection. The negatives are the reason to trust the rest.
Each paper names, in its own limitations section, the result that would prove it wrong. That is unusual and it is deliberate. Several also correct earlier versions of themselves in the open rather than quietly revising, including one that retires an explanation I had published in an earlier paper.
Every paper is sole-authored by Ethan Bradley, ORCID 0009-0008-8925-7975. Once each preprint is posted it receives a permanent identifier, and this page will carry it. Until then the identifiers are marked pending rather than estimated.
If you find an error in any of them, I would rather hear it than not: ethan@brugada.net.