Experimental protocols, costed and staged for laboratory execution.
What began as two open questions is now a staged, mathematically bounded experimental design. Both protocols are designed to be falsifiable, and a clean negative result would be published as readily as a positive one.
This is not a request about anyone's care.
I carry SCN5A p.Arg104Gln and live with Brugada syndrome. My clinical care rests with my physicians. These protocols exist because the questions they answer decide whether four therapeutic routes are open or closed, affecting families who carry variants in this region, several of whom do not know it yet.
Construct provenance: why papers do not share a single Wild-Type
A seven-study matrix across the SCN5A literature revealed that published experiments do not use an interchangeable wild-type channel. Backgrounds differ at exon 6 (adult versus neonatal), glutamine 1077 (Q1077del versus Q1077ins), the common H558R polymorphism, and whether the beta-1 auxiliary subunit is co-expressed. Any comparison that ignores construct background risks mistaking an artifact of plasmid cloning for a property of the variant.
Tagging introduces a second distinction. Cell-surface assays frequently tag Nav1.5 with Green Fluorescent Protein at the N-terminus. The 1.41-million-atom molecular dynamics production run verified that this tag does not collapse the DEKA selectivity filter (filter radius remains 5.23 angstroms with 12 hydrating waters). However, because fusion tags can affect trafficking kinetics, untagged native peptides remain the required standard for quantitative turnover.
The mass spectrometry discovery: Lys-C versus Trypsin
The mechanism experiment requires counting wild-type and mutant channels separately at the cell surface. Antibodies cannot distinguish them because the single letter change is buried in the cytoplasmic fold. The solution is enzymatic digestion and mass spectrometry.
Standard trypsin cuts after arginine. Because position 104 is itself an arginine, trypsin destroys the site on the wild-type copy, creating an undetectable 4-amino-acid fragment while leaving a 21-amino-acid mutant fragment. The two cannot be quantified as a ratio because the mutation deletes the cleavage site.
Endoproteinase Lys-C cuts exclusively after lysine, leaving position 104 untouched. Both copies yield an identical 26-residue peptide fragment differing by exactly 28.04 Daltons in molecular weight (arginine 156.19 versus glutamine 128.15), easily resolved on standard liquid chromatography-tandem mass spectrometry instruments.
The staged decision tree: conditional branches
Rather than funding a monolithic program all at once, the protocol proceeds in staged, decision-gated branches:
The cost is the cells, not the measurement
Building and validating the cell lines accounts for 72 percent of the mechanism budget (15,500 dollars). Once the cell substrate exists, individual measurements cost between 4,000 and 7,000 dollars. A funder paying for one experiment answers one question; a funder supporting the stable cell lines establishes the physical substrate for all four stages.
Sample sizes on the corrected comparator
Sample sizes for the non-inferiority arms were recalibrated against the corrected 45.8 percent wild-type baseline (from O'Neill's measured 2.184 two-allele scaling factor). At statistical margins of 12.5, 10, 7.5, and 5 percentage points, required sample sizes are 63, 98, 174, and 390 cells per group. Cell-to-cell variability across deposited automated patch datasets ranges from 57 to 89 percent in cell lines, and 18 to 25 percent in primary human cardiomyocytes, demonstrating that preparation quality dictates statistical power.
The full protocol specifications, line-item spreadsheets, primer sequences, and decision trees are available on request. Write to ethan@brugada.net.