[us-commits] [ehb54/ultrascan3] fa2ad0: perceiver: phosphate convention for residues with ...

[emre brookes]

  Branch: refs/heads/ehb54-issue-980
  Home:   https://github.com/ehb54/ultrascan3
  Commit: fa2ad021f3ad467403b337a3166ef384eb0242c0
      https://github.com/ehb54/ultrascan3/commit/fa2ad021f3ad467403b337a3166ef384eb0242c0
  Author: ehb54 <brookes at uthscsa.edu>
  Date:   2026-08-12 (Wed, 12 Aug 2026)

  Changed paths:
    M us_somo/develop/include/us_hydrodyn_perceive_dialog.h
    M us_somo/develop/perceiver/tests/tests_unit.cpp
    M us_somo/develop/src/us_hydrodyn_perceive.cpp
    M us_somo/develop/src/us_hydrodyn_perceive_dialog.cpp
    M us_somo/somo/doc/manual/somo/somo_perceive.html

  Log Message:
  -----------
  perceiver: phosphate convention for residues with >2 ionizable groups

somo.residue holds at most two ionizations per residue, so citrate's three
carboxyls used to collapse the whole entry to its pH 7 state -- correct at
pH 7 and frozen everywhere else.

Adopt the convention the coded table already uses for phosphate (ehb54 and
Mattia, 2026-08-12: "assume the first oxygen to be always deprotonated").
The lowest-pKa groups are written permanently deprotonated in the primary
fields; the two highest keep real switchable slots. Lowest-first is what
makes it defensible: that group is already fully ionized wherever the entry
is usable. Phosphate (2.15/7.20/12.35) reproduces the hand-written entry
exactly; citrate keeps its 2-/3- transition instead of being frozen.

The pKa becomes per-group rather than per-residue -- two switchable groups
can differ (citrate 4.76 and 6.40) and the header stores one pair per
ionization, selected by the atom's ionization index. One predicate now
drives both the atom line and the header pair so an index can never
reference a pair that was not declared.

Also fixes the bead hydration, which summed the protonated column while the
coded residues count ionizable atoms ionized (Asp's OD2 is 0/5 and Asp's
second bead is written 5). 2CMD's citrate emitted a bead of 1 against the 16
its own review dialog computed for the same entry; both now say 16.

6 new unit tests; regression unchanged at 99.833% geometric perception.
Help page updated: per-group pKa, and the convention documented.

Co-Authored-By: Claude Opus 5 <noreply at anthropic.com>



To unsubscribe from these emails, change your notification settings at https://github.com/ehb54/ultrascan3/settings/notifications