Branch: refs/heads/ehb54-issue-980 Home: https://github.com/ehb54/ultrascan3 Commit: bb17af1bcaf6164b068ae745e30cca99b1222cb1 https://github.com/ehb54/ultrascan3/commit/bb17af1bcaf6164b068ae745e30cca99b... Author: ehb54 <brookes@uthscsa.edu> Date: 2026-08-12 (Wed, 12 Aug 2026) Changed paths: M us_somo/develop/src/us_hydrodyn_hydration.cpp M us_somo/develop/src/us_hydrodyn_perceive_dialog.cpp Log Message: ----------- perceiver: readable table headers, and stop flagging plain carbonyl oxygens Two things from Mattia's screenshot of 8RAT_gap3 / LYZ. Headers unreadable: QHeaderView paints its sections with Button/ButtonText, not Window/WindowText. perceive_palette() set only the text, so the section background stayed inherited and the headers came out dark on dark. Button is now an explicit light grey, and the palette is applied to the table and its header widget -- neither inherits the frame's -- with the header in bold so it reads as one. Spurious review flag: an oxygen with no hydrogen that is not part of a carboxyl -- a backbone carbonyl, an ester, an ether, a phosphate O -- matched no rule and fell through to "no pH 7 rule for this group", so nearly every entry was flagged for its own backbone. Every O1H0 in somo.residue carries 0 waters (36 backbone O, plus O2/O4/O6/OP1/OP2 and the rest), so that is a confident zero and is now stated as one. LYZ in 8RAT_gap3 goes from 3 flagged items to 1, the remaining one being the real ASA placeholder note. Tests unchanged: unit 86/0, hydration 37/0, psv 60/0. libus_somo builds clean. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Commit: 71f1ee61a3daf2baf4560952ef17abf67f8c6c8e https://github.com/ehb54/ultrascan3/commit/71f1ee61a3daf2baf4560952ef17abf67... Author: ehb54 <brookes@uthscsa.edu> Date: 2026-08-12 (Wed, 12 Aug 2026) Changed paths: M us_somo/develop/src/us_hydrodyn_core.cpp M us_somo/develop/src/us_hydrodyn_perceive_dialog.cpp Log Message: ----------- perceiver: show both hydration totals, and point at the module from the load Two from Mattia's testing. Hydration total read 1 for 2CMD. That is the PROTONATED total and it is correct as far as it goes -- since ionizable atoms started carrying their waters in the deprotonated alternate, citrate's three carboxyls contribute 0 each and only the central hydroxyl's 1 remains. But at pH 7 citrate is deprotonated, so the number that matters is 16, and showing only the neutral one is misleading. The label now gives both, and only when the entry actually has an ionizable atom: Residue hydration total: 1.00 waters neutral, 16.00 ionized Missing perceiver warning: no such message has ever existed in this branch -- I searched the history for one and there is none -- so this is a gap rather than a regression, but the gap is real. The load reports non-coded residues, the bead builder then represents each by a single crude bead, and nothing told the user an entry could be derived from the coordinates first. check_for_missing_atoms now says so once unknown_residues is final, naming the residues it means. libus_somo builds clean. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Compare: https://github.com/ehb54/ultrascan3/compare/8b715d0c3744...71f1ee61a3da To unsubscribe from these emails, change your notification settings at https://github.com/ehb54/ultrascan3/settings/notifications