[us-commits] [ehb54/ultrascan3] 8e3c1b: extrap_c0: optional post-fit SD reassessment of th...

[emre brookes]

  Branch: refs/heads/ehb54-issue-920
  Home:   https://github.com/ehb54/ultrascan3
  Commit: 8e3c1bee65b5781447da92afc9f81e9363b8000b
      https://github.com/ehb54/ultrascan3/commit/8e3c1bee65b5781447da92afc9f81e9363b8000b
  Author: ehb54 <brookes at uthscsa.edu>
  Date:   2026-07-13 (Mon, 13 Jul 2026)

  Changed paths:
    M us_somo/develop/include/us_hydrodyn_saxs_iqq_extrap_c0_conc.h
    M us_somo/develop/src/us_hydrodyn_saxs_iqq_extrap_c0.cpp
    M us_somo/develop/src/us_hydrodyn_saxs_iqq_extrap_c0_conc.cpp
    M us_somo/somo/doc/manual/somo/somo_saxs_extrap_c0.html

  Log Message:
  -----------
  extrap_c0: optional post-fit SD reassessment of the extrapolated curve

Add a "Recompute output SDs" drop-down to the extrapolation concentration
dialog (Off by default) that reassesses the extrapolated curve's error bars
after the fit, reusing the HPLC/KIN s.d.util "SD rescale" engine
(US_Saxs_Util::recompute_errors, BayesApp-style; Larsen & Pedersen 2021).

Modes: Constant (one factor for all q), Non-constant (per q-bin), and
Intensity-dependent (sd -> sd + a*|I|); Off keeps the propagated intercept
errors. The reassessment runs on the final output curve (out_I0/out_q/
out_I0_err), so when a reference tail is spliced in above the merge q it
spans the whole spliced curve. sigma is rescaled only when the fit warrants
it (poor fit and >10% off unity); the verdict, reduced chi^2 and p-value are
logged (p fixed-point, matching the s.d.util convention). Every recompute_errors
false-return path precedes any write to sd, so a skipped reassessment leaves
the propagated errors intact.

The chosen mode is encoded in the self-describing curve name
(_sdC / _sdN / _sdI). Manual updated (modes section + name tokens).

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



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