[hyphaltip]

Docs only. This adds docs/notes/2026-11-09_junction-panel-design.md, a design proposal the curator asked for on 2026-11-09 so that #195 and #196 can be ruled on and v0.9.0 can close. No analysis was run; the numbers come from the committed cap-700 wave tables on main.

The note covers:

1. What a junction panel types.
   • Silent cassettes share Y and its borders with the expressed locus, so only sequence unique to that locus separates them.
   • It proposes A, a copy-number ratio test first. With standard cassettes Ya:Yalpha is 2:1 for MATa, 1:2 for MATalpha and 1:1 for a/alpha. It works from the depth_1 and depth_2 the panels already report.
   • It proposes B, a linked-pair junction panel (Y plus the MAT-unique outer flank), only if A fails the controls.
2. "Cannot be typed" (Reads: species where presence of an idiomorph cannot be typed (silent cassettes, homothallic, switching) and what the gate should say #195). One gate reason per cause: homothallic_locus, silent_cassettes, zygosity (exists) and cause_unchecked, each with draft site wording.
   • It points out that K. marxianus, K. phaffii and S. kudriavzevii are cassette yeasts currently shown as "provisional (homothallic)".
   • It points out that S. uvarum's single calls (263 of 294 runs) are unexplained.
3. Yeast panels (Reads: yeast panels (Saccharomyces sensu stricto, Candida albicans, Zygosaccharomyces, Lachancea and others) withheld in wave 2 #196). Species by group, with the first step for each. Truth sets: simulated mixes, S. cerevisiae BY4741/2/3, S. pombe 972/975/968, published C. albicans MTL zygosity, and S. sclerotiorum as the homothallic control.
4. Done and not done. What is done, plus 8 remaining steps with dev and compute cost each.
5. Six open questions for the curator. The note suggests v0.9.0 closes on the gate reasons and wording, the cause TSV and C. fimbriata. The ratio and junction typing would move to v0.10.

Full suite on the HPCC at 50e43ac: 1962 passed, 6 skipped.

🤖 Generated with Claude Code