SCN5A R104Q: guide design, off-target scan, and ClinVar functional-evidence census

Superseded, 6 August 2026. A corrected version 2 of this deposit has been published at 10.5281/zenodo.21840036. One file in this version is wrong: MS_TABLE3_LEAD_PROTEIN_CHANGING.csv holds 16 data rows, while the paper that cites it states the count is 22 and prints 22 in its own Table 3. The paper is right; this file was built from a superseded consequence column and was never rebuilt after the recount. A second file in this version is also defective, and it is defective in version 2 as well: ABE_CONSEQUENCE_RECHECK.csv truncates its recomputed_aa field at exactly 1,500 characters, so 58 of its 668 call-carrying rows end mid-token and 1,706 of the 8,843 transcript-level consequence calls it should carry, 19.3 percent, are absent. The 22,717-row count is right and says nothing about the defect. That file has not been repaired in either version, because the script that produced it no longer exists; the missing text is unrecoverable and a regenerated table is the only clean fix. A third defect is an absence rather than a bad file: this version contains no data file for paper 10 (10.5281/zenodo.21799871) at all, although that paper's data availability statement names four tables as deposited here. Version 2 adds seven paper-10 files, three of them regenerated from public coordinates and marked as reconstructions, and paper 10's version 2 rewrites the statement to describe what is actually present. The other 50 files in this version are byte-identical to version 2. This version remains available and citable for what it is; if you are checking a paper's off-target count against this deposit, use version 2, and read section 3.4.8 of paper 2 before reusing the recheck table for anything. Data tables behind two preprints on adenine base editor guide design for SCN5A p.Arg104Gln and on the retrievability of deposited functional evidence in ClinVar. Computational predictions and database measurements only; no experimental validation.